Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 27.
Published in final edited form as: Psychol Assess. 2017 Dec 4;30(6):719–728. doi: 10.1037/pas0000521

Evaluating the 8-Item Patient Health Questionnaire’s Psychometric Properties with Mexican and Central American Descent University Students

David Alpizar 1, Luciana Lagana 2, Scott W Plunkett 3, Brian F French 4
PMCID: PMC7591264  NIHMSID: NIHMS963819  PMID: 29199838

Abstract

Depression can be a serious aliment influencing the lives of millions of persons. Person’s health conditions worsen and daily functioning declines in the presence of depression. However, little attention has been given to how depression is accurately assessed in diverse populations from other countries residing in the United States. Thus, this study examined the factor structure, factorial invariance, concurrent, and convergent validity of the widely-used 8-item Patient Health Questionnaire (PHQ-8) among emerging adults of Mexican, Central American (e.g., Salvadorians, Guatemalans) men and women residing in the United States. Undergraduate student participants (n = 2782) completed an online or paper-pencil version of the assessment. Support was found for 2-factor and 1-factor solutions, however the unidimensional structure was recommended due to the very high correlation between the two factors (r ≥ .87). The unidimensional scale was invariant between (1) Mexican and Central American descent and (2) men and women. Depressive symptoms (as measured in the PHQ-8) were significantly correlated with another valid and reliable measured of depression, and scores on the PHQ-8 were significantly correlated with theoretically related variables (i.e., quality of life, parental support, and perceived stress).

Keywords: Depression, PHQ, Latina/o, college students, emerging adults

Depression is characterized by depressed/irritable mood, absence of interest or pleasure, change in appetite and sleep routines, fatigue, feelings of guilt/worthlessness, problems concentrating, and suicidal thoughts (American Psychiatric Association [APA], 2013). Depression is a disorder that can impact a person’s well-being and daily functioning (Oquendo, Lizardi, Greenwald, Weissman, & Mann, 2004) and worsen physical health such as cardiovascular disease (American Heart Association, 2013). Also, depression has comorbidity with other mental disorders (e.g., substance abuse disorders, Lehmann, Hubbard, & Martin, 2001) and can potentially lead to suicide ideation (Oquendo et al., 2004). Hence, depression decreases the overall well-being of the affected individuals.

University students are at-risk for developing depression because they face challenges at school (Eisenberg, Gollust, Golberstein, & Hefner, 2007) as well as in the multiple transitions associated with emerging adulthood such as finding romantic partners, identity exploration, moving, and finding a job (Arnett, 2011; Adkins, Wang, Dupre, Oord, & Elder, 2009; Hill, Jackson, Roberts, Lapsley, & Brandenberger, 2011). Another high-risk group for depression is Latina/os (Gonzalez et al., 2010), including Latina/o college students (Granillo, 2012). However, conflicting results regarding rates of depression of individuals of Central American descent and Mexican descent exist (Camacho et al., 2005; Salgado de Snyder, Cervantes, & Padilla, 1990). Regardless of these differences, there is a need for depression inventories with appropriate psychometric evidence among emerging adults of Mexican and Central American descent.

The Patient Health Questionnaire (PHQ-9; Spitzer et al., 1999) is a widely used 9-item measure, which assesses depressive symptoms in individuals. The PHQ-9 has become widely used for screening for depression as well as research on depressive symptoms (Kroenke, Spitzer, Williams, & Löwe, 2010) because the measure (1) has items that directly correspond to the depression diagnostic criteria in the DSM-V (APA, 2013); (2) is free, publicly available, short, and quick to administer and score (Spitzer et al., 1999); and (3) has good psychometric properties in different settings (e.g., clinical, research, school; Kroenke & Spitzer, 2002) and in different countries (see meta-analysis by Moriarty, Gilbody, McMillan, & Manea, 2015). Various short forms of the PHQ-9 have been developed, such as the PHQ-8 (8 items) and the PHQ-2 (2 items). The PHQ-8 does not include the 9th item regarding suicidality. The current study uses the PHQ-8 over the PHQ-9 for the following reasons. First, studies comparing the PHQ-8 and PHQ-9 have found extremely high correlations (i.e., > .997; Kroenke & Spitzer, 2010; Razykov, Ziegelstein, Whooley, & Thombs, 2012) and virtually identical scoring thresholds for depression (Kroenke & Spitzer, 2002, 2010) between the two measures. Kroenke and Spitzer (2010) found the 9th item does not discriminate well between depressed and not depressed respondents. Thus, the PHQ-8 and PHQ-9 scores result in the same substantive conclusions. Also, studies with Latina/o samples found the 9th item generally had the smallest factor loading (e.g., Familiar et al., 2014; Huang et al., 2006; Merz et al., 2011). And finally, researchers may have difficulty getting their survey approved by an institutional review board when including an item about suicidality.

There are mixed findings about the factor structure of the PHQ-8. The PHQ-8 does have support for being unidimensional among Latina/os (Huang, Chung, Kroenke, Delucchi, & Spitzer, 2006; Merz, Malcarne, Roesch, Riley, & Sadler, 2011) and men of Mexican descent (Donlan & Lee, 2010). However, the unidimensionality of the PHQ-8 is not fully established, given evidence supporting two factors (i.e., affective and somatic) with a Latina college sample after removing two items (i.e., Granillo, 2012). Due to these inconsistencies, additional evaluation of the PHQ-8’s factor structure and invariance across Latina/o groups is warranted. Otherwise, levels of depressive symptoms may be over or underestimated in different populations (Chen, 2008).

In addition to establishing internal structure validity evidence within and between groups, it is important to examine validity evidence focused on associations with other variables. Specifically, convergent validity is the relationship between a tested inventory and a previous validated inventory documented to measure the same construct in the literature. In this case, the PHQ-8 should be examined with a valid depression inventory to support that it assesses the same construct. Second, it is important to evaluate the concurrent validity of the PHQ-8. Concurrent validity is the degree to which an inventory relates to theoretically related variables. The following variables have been documented to relate to depression in the literature: perceived support from parents, perceived stress, and quality of life. Parental support refers to the amount of involvement and encouragement that parents provide to their offspring (Peterson, 2005). Scholars have documented that emerging adults who receive more parental support have lower levels of depressive symptoms (e.g., Liem, Lustig, & Dillon, 2010; Needham, 2008). Additionally, perceived stress is defined as an unexpected situation/social challenge (e.g., car accident) where negative feelings/behaviors (e.g., anger, irritation) are produced (Ng & Jeffery, 2003). Adults who experienced stressful situations have higher symptoms of depression (Flores, Tschann, Dimas, Bachen, Pasch, & de Groat, 2008; Ng & Jeffery, 2003), including Latina/os (Vega, Rodriguez, & Ang, 2010). Quality of life corresponds to the overall well-being (e.g., mental, physical, social, and emotional health) of an individual (Husson, Mols, & Van de Poll-Franse, 2010). Several studies have shown that higher levels of depression negatively impact well-being (Abraham & Fava, 1999; Gotlib, 1997). When individuals are depressed, they find less pleasure in activities, experience more physical fatigue, and have lower self-worth, all of which can undermine quality of life. Because the importance of parental support, perceived stress, and quality of life are widely documented in the literature, the relationship of these theoretically related variables with depression should be examined to provide validity evidence.

Statement of the Problem

Depression can be detrimental to an individual’s well-being (Abraham & Fava, 1999; Gotlib, 1997; Oquendo et al., 2004). The 8-item Patient Health Questionnaire (PHQ-8) is a self-administered instrument used to assess mental disorders; it is widely used in research and clinical practice to measure depression (Kroenke et al., 2010). However, there is a dearth of information about how the scores function with Mexican American and Central American descent university students residing in the United States. Only a single study has targeted Mexican descent emerging adults, and most studies have grouped Mexican and Central American descent with other Spanish speaking groups as Latina/os or Hispanics. There is available evidence that different Latina/o groups report differing levels of depression (e.g., Oquendo et al., 2004), thus it is beneficial to separate Latina/o groups originating from different regions (e.g., Mexico, Central America) when studying depression. In addition, depression inventories’ psychometric properties should be examined across these groups. If unreliable and invalid scores on these scales are used to measure depressive symptoms, they may not capture the desired construct and could lead to inaccurate findings. Thus, the purpose of this study was to examine the PHQ-8’s psychometric properties among university students of Mexican and Central American descent. Since women (i.e., Latinas) have reported higher rates of depression compared to men (Brown, Meadows, & Elder, 2007), differences between men and women were also examined.

To be more specific, the current study investigated the internal structure (i.e., factor structure, factor invariance) of the PHQ-8 and association with other variables (e.g., criterion validity) across the target populations. The current study has the following hypotheses:

  1. Hypothesis 1 aimed to establish the factor structure of the PHQ-8. The literature has provided mixed findings about the factor structure (unidimensional vs. multidimensional) of this inventory. Therefore, the factor structure was examined across Mexican and Central American descent, as well as men and women. For hypothesis 1, it was hypothesized the PHQ-8 items would support the general depression factor for each subsample in accordance with how the scale was initially developed and then modified by Spitzer et al. (1994, 1999).

  2. Hypothesis 2 aimed to test invariance (i.e., configural, metric, and scalar) of the PHQ-8 for (1) Mexican American and Central American descent and (2) men and women. The current study hypothesized the PHQ-8 would meet factor invariance (i.e., configural, metric, and scalar) for each subsample.

  3. Hypothesis 3 aimed to examine the concurrent and convergent validity of the PHQ-8 by examining correlations between PHQ-8 scores and other variables for (1) Mexican American and Central American descent and (2) men and women. Specifically, the depression factor was expected to relate positively to a previous validated depression inventory (i.e., Center for Epidemiological Scale of Depression) in each subsample. Also, for each subsample, the depression factor was expected to correlate positively to perceived stress. Finally, the depression factor was expected negatively to perceived parental support and quality of life in each subsample.

Methods

Procedures and Sample

This study was approved by the university’s Institutional Review Board (IRB). The participants were recruited from a psychology subject pool across three semesters at a university in the greater Los Angeles area of the United States. The participants came from a lower-division, general education course (i.e., introductory psychology). Participation in research studies or completing an alternative assignment was required for students in the course. In two semesters, participants who signed up for the study completed an online survey. In one semester, participants who signed up for the study completed a paper-pencil survey in a classroom setting (10–20 students in a group). Trained research assistants (RAs) distributed the survey to students. Different semesters included different variables in the study, thus sample size are different depending on variable in the analyses. The current sample only uses data from participants who ranged from 18 to 29 years of age, self-reported as Mexican or Central American descent, and identified as women or men. Table 1 provides the demographic characteristics of the subsamples.

Table 1.

Demographic Characteristics of the Sample

Mexican Descent Central American Descent Men Women
N 2034 748 802 1980
Age (Range is 18–29 years)
 M 19.3 19.3 19.4 19.2
 SD 1.6 1.7 1.7 1.6
Sex
 Men 28.8% 28.9%
 Women 71.2% 71.1%
U.S Born
 Yes 88.3% 88.9% 90.1% 88.2%
 No 10.7% 11.1% 9.9% 11.8%
Parents’ Marital Status
 Married 66.5% 59.2% 67.1% 63.6%
 Divorced 9.9% 8.3% 8.9% 9.7%
 Never married 8.1% 11.9% 8.9% 9.2%
 Remarried 2.7% 4.1% 3.0% 3.1%
 Widowed 3.0% 2.1% 2.1% 3.1%
 Separated 5.9% 10.4% 6.7% 7.3%
 Other 1.9% 2.5% 1.7% 2.2%
 Missing 1.9% 1.2% 1.6% 1.8%
Reside with Parent
 Yes 56.0% 55.5% 60.0% 54.2%
 No 19.7% 16.0% 16.5% 19.6%
 Missing 24.3% 28.5% 23.6% 26.2%
Classification
 Freshman 52.5% 53.1% 52.7% 52.6%
 Sophomore 28.4% 25.4% 26.9% 27.8%
 Junior 12.5% 14.0% 12.8% 12.9%
 Senior 6.4% 7.4% 7.4% 6.5%
 Masters 0.1% 0.1% 0.0% .2%
 Missing 0.1% 0.0% .2% 0.0%
Generation Status
 1st generation (all foreign born) 11.6% 10.0% 9.7% 11.8%
 2nd generation (subject USA born, parents foreign born) 75.1% 83.3% 78.2% 76.9%
 2.5 generation 10.7% 6.3% 10.0% 9.3%
 3rd generation (all USA born) 2.7% .4% 2.1% 2.0%
Open in a new tab

Measurement

A protocol was developed that included a demographics questionnaire (e.g., gender, age, ethnicity, birth country of participant and parents). The primary variables were measured with scales that have been shown to be reliable (internal consistency) and valid in other studies with Latina/os and/or college students. For each of the scales, the items were averaged to create a scale score.

Patient Health Questionnaire (PHQ-8)

Depressive symptoms were measured with the 8-item PHQ (Spitzer et al., 1999). The stem for the items was: “How often have you been bothered by each of the following symptoms during the past two weeks?” Two sample items follow: “Feeling down, depressed, irritable, or hopeless” and “Little interest or pleasure in doing things.” The response choices were: 0 = not at all, 1 = several days, 2 = more than half the days, and 3 = nearly every day. The literature has documented internal consistency reliability for Latina/os (e.g., α = .80, Huang et al., 2006), including Mexican individuals (i.e., α = .91, Donlan & Lee, 2010).

The Center for Epidemiological Scale – Depression (CES-D) Short Form

Depressive symptoms were also assessed with the 10-item short form (Kohout, Berkman, Evans, & Cornoni-Huntley, 1993) of the 20-item CES-D (Radloff, 1977). The stem for the items follows: “How often have you felt the following ways during the past SEVEN DAYS.” Sample items were: “I felt sad” and “I enjoyed life” (reverse coded). Two items were reverse-coded. The response choices were: 0 = rarely or none of the time (less than1 day), 1 = Some or a little of the time (1–2 days), 2 = Occasionally (3–4 days), and 3 = mostly or almost all the time (5 to 7 days). The CES-D has been used in research and clinical assessment in order to measure depression (Crockett et al., 2005). The 10-item CES-D has been shown to have good factor structure and reliability with Mexican American migrant and immigrant adults (Grzywacz, Hovey, Seligman, Arcury, & Quandt, 2006). In the current study, the following Cronbach’s alphas were found: .86 for Mexican American, .84 for Central American descent, .85 for men, and .85 for women.

Perceived parental support

Perceptions of support from mothers and fathers were evaluated using the 4-item maternal and paternal subscales of the Parental Behavior Measure (Bush, Peterson, Cobas, & Supple, 2002). Two sample item follows: “This parent has made me feel that he/she would be there for me if I needed him” and “This parents tells me how much she/he loves me” Each item was asked twice, assessing the ratings of support from mothers and fathers separately. The response choices were: 1 = strongly disagree, 2 = disagree, 3 = agree, and 4 = strongly agree. Good internal consistency reliability (α > .78) has been established for Latina/o adolescents (Behnke, Plunkett, Sands, & Bámaca-Colbert, 2011; Umaña-Taylor & Guimond, 2012). The items for mother and father were averaged together. In the current study, the following Cronbach’s alphas were found: .86 for Mexican American, .86 for Central American descent, .88 for men, and .86 for women.

Perceived stress

Emerging adults’ perceptions of uncontrollable, unpredictable, and overloading life events were measured utilizing the 10-item Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983; Cohen & Williamson, 1988). Two sample items follow: “Been upset because of something that happened unexpectedly?” and “Felt nervous and ‘stressed’?” The response choices were: 0 = never, 1 = almost never, 2 = sometimes, 3 = fairly often, and 4 = very often. Four items were reversed coded. Strong reliability and validity evidence for the scores has been established across different ethnic groups (Flores et al., 2008; Roberti, Harrington, & Storch, 2006). Previous studies provided internal consistency reliability estimates for a Latina/o population, with a Cronbach’s alpha of .70–.77 for English-speakers and .75 for Spanish-speakers (Flores et al., 2008; Zambrana, Scrimshaw, Collins, & Dunkel-Schetter, 1997). In the current study, the following Cronbach’s alphas were found: .85 for Mexican American, .85 for Central American descent, .84 for men, and .85 for women.

Quality of life

Emerging adult’s overall quality of life was measured with a single-item (Zimmerman et al., 2006). Participants were asked “In general, how would you rate your overall quality of life during the last six months?” The response choices follow: 0 = Very bad, my life could hardly be worse, 1 = Pretty bad, most things are going poorly, 2 = The good and bad parts are about equal, 3 = Pretty good, most things are going well, and 4 = Very good, my life could hardly be better. Zimmerman et al. reported high test-retest reliability (intraclass r = .81). Specifically, depressed patients reported significantly lower quality of life than non-depressed patients, and quality of life was significantly related to a range of theoretically related variables (Zimmerman et al., 2006).

Results

Analyses

The current study collected sources of evidence to build a validity argument for the inferences and uses of scores from the PHQ-8. Under the theoretical validity framework (see Kane, 2013), there are many sources of validity evidence (e.g., internal structure, response process) that can be used to support a validity argument. We focused on internal structure (i.e., hypothesis 1 and hypothesis 2) and association with other variables (i.e., hypothesis 3) for target sources of validity evidence.

To test hypothesis 1, a series of confirmatory factor analysis (CFA) models, via Mplus, were used with Mexican, Central American, men, and women participants. Testing multiple models provides stronger validity evidence (Thompson & Daniel, 1996). The current study examined a 1-factor (i.e., unidimensional) and 2-factor structure (i.e., multidimensional), consistent with prior reports. Given the PHQ-8 items were categorical, weighted least squares means and variance (WLSMV) were employed as recommended by Finney and DiStefano (2013). It is worth mentioning that degrees of freedom and chi-square for WLSMV estimation is an adjustment to obtain accurate p-values (see Mplus user’s appendix 4), leading to some values that may appear inaccurate given standard calculations of degrees of freedom.

The model fit was assessed using the χ2, the Comparative Fit Index (CFI ≥. 95) and the standardized root mean square residual (SRMR < .08). These cutoffs were recommended in the SEM literature (Hu & Bentler, 1999; Tabachnick & Fidell, 2012). Meeting these criteria indicated acceptable fit. In the literature, less stringent criteria for fit indexes (e.g., CFI > .90) are recommended since fit indices lack clarity of their performance with WLSMV estimation (e.g., Finney & DiStefano, 2013). Yet, the current study opted for more stringent criteria for providing stronger support for the internal structure.

In hypothesis 2, multi-group confirmatory analyses were used to test factor invariance across (1) Mexican and Central American descent and (2) men and women. To test the factor invariance between the groups, the following steps took place. First, configural invariance, which was the same factor structure between the groups, was evaluated. Also, parameters (e.g., factor loadings) were free between the models. Second, metric invariance, where equality constraints were applied on the loadings, was evaluated. Also, the metric step was tested against the configural step. Finally, scalar invariance, where equality constraints were applied on loadings and thresholds, was evaluated. Also, the scalar step was tested against the configural step and then against metric. The model fit was evaluated in each step for invariance testing. As recommended by the literature (Hu & Bentler, 1999; Tabachnick & Fidell, 2012), acceptable model fit was assessed: χ2, CFI ≥. 95. However, to judge if the model equality constraints to assess invariance worsened fit, the difference in χ2 (Δχ2) was used (French & Finch, 2006) in combination with the change in CFI (ΔCFI; Chen, 2007) greater than 0.01. This combination of indices guards against the claim that the use of one index (e.g., Δχ2) be over sensitive to model differences and sample size. Thus, to indicate a lack of invariance, the Δχ2 had to be significant and the ΔCFI had to be greater than 0.01.

To test hypothesis 3, Pearson correlations tested the relation of the PHQ-8 with other variables (i.e., concurrent and convergent validity) for Mexican descent, Central American descent, men, and women. First, we examined the concurrent criterion by examining the association between the PHQ-8 and the CES-D. Second, we examined the criterion validity by evaluating the relationship between the PHQ-8 and theoretically related variables (i.e., stress, perceived parental support, and quality of life). To evaluate the significance of the correlation analyses, the current study used α = .01. Evaluating the obtained correlation coefficient was based on the following criteria (see Cohen, 1988): (1) r < .10 was considered a low correlation, (2) r > .30 and r < .50 was a medium correlation, and (3) r > .50 was considered a strong correlation.

Data Screening

Data screening was conducted by group (i.e., Mexican, Central American, men, and women). The PHQ-8 items had less than 5% missing values in each subsample. Little’s MCAR test suggested data were missing at random for the Mexicans (χ2(26) = 49.156, p > .01), Central Americans (χ2(30) = 42.666, p > .01), women (χ2(27) = 51.354, p > .01), and men (χ2(22) = 28.115, > .01). Next, no outliers were detected with z-score great than 3.3 (Tabachnick & Fidell, 2012). Also, no concerns related to skewness and kurtosis factor or scale means (i.e., skew and kurtosis values were below the absolute value of 2 for both indices). To minimize the concerns, the literature has recommended using robust methods (i.e., bootstrapping, Field, 2013), which we employed to take a conservative approach.

Results

The polychoric correlation matrices are presented in Table 2 for Mexican and Central American descent and for men and women.

Table 2.

Polychoric Correlation Matrix for Each Group

PHQ Items Mexican (below), Central American (above) Men (Below), Women (Above)
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
Feeling down, depressed, irritable, or hopeless? -- .71 .48 .54 .58 .68 .55 .58 -- .73 .51 .54 .57 .70 .60 .59
Little interest or pleasure in doing things? .74 -- .57 .57 .63 .63 .62 .56 .74 -- .57 .53 .58 .62 .61 .58
Trouble falling asleep, staying asleep, or sleeping too much? .53 .54 -- .63 .69 .41 .58 .49 .53 .49 -- .64 .63 .44 .56 .53
Poor appetite, weight loss, or overeating? .56 .51 .64 -- .65 .56 .58 .61 .56 .52 .63 -- .63 .52 .56 .56
Feeling tired, or having little energy? .54 .55 .60 .62 -- .50 .63 .51 .50 .53 .61 .62 -- .50 .58 .49
Feeling bad about yourself – or feeling that you are a failure, or that you have let yourself or your family down? .74 .62 .49 .53 .51 -- .64 .64 .77 .62 .51 .59 .53 -- .63 .62
Trouble concentrating on things such as schoolwork, reading, or watching TV? .61 .58 .56 .56 .56 .64 -- .61 .59 .53 .56 .60 .59 .67 -- .60
Moving or speaking so slowly that other people could have noticed?...Or the opposite – being so fidgety or restless that you were moving around a lot more than usual? .61 .59 .54 .57 .52 .64 .62 -- .62 .58 .54 .64 .59 .69 .65 --
Open in a new tab

Note: All correlations were significant at least p < .05.

Validity Evidence for Internal Structure

Hypothesis 1 evaluated the PHQ-8 factor structure for Mexicans, Central Americans, men, and women. The 1-factor model met fit criteria for Mexicans (χ2(20) = 517.404, p < .05, CFI = .969, SRMR = .048), Central Americans (χ2(20) = 220.064, p < .05, CFI = .967, SRMR = .057), men (χ2(20) = 233.433, p < .05, CFI = .966, SRMR = .051), and women (χ2(20) = 482.749, p < .05, CFI = .970, SRMR = .049). Also, the 2-factor model met fit criteria for Mexicans (χ2(19) = 279.294, p < .05, CFI = .984, SRMR = .036), Central Americans (χ2(19) = 128.163, p < .05, CFI = .982, SRMR = .045), men (χ2(19) = 134.775, p < .05, CFI = .982, SRMR = .038), and women (χ2(19) = 265.228, p < .05, CFI = .984, SRMR = .037).

The standardized loadings in both models were significant (see Table 3) and above .70 for Mexicans, Central Americans, men, and women. However, the 2-factor model had practical concerns. The latent variables were significantly (p < .05) and highly correlated (i.e., Mexican descent, r = .875; Central American descent, r = .870; men, r = .874; and women, r = .878). Highly correlated latent variables were not representing a distinction between constructs as recommended (e.g., Kline, 2005). Thus, the 1-factor model was retained. Also, the 1-factor model is consistent and practical for clinical and research settings, and the 1-factor model had approximately equal internal consistency estimates for Mexicans (i.e., α = .88), Central Americans (i.e., α = .89), men (i.e., α = .89), and women (i.e., α = .88).

Table 3.

Standardized Factor Loadings of the Confirmatory Factor Analyses

PHQ items Mexican Descent Central American Descent Male Participants Female Participants
1-factor 2-factor (r = .875) 1-factor 2-factor (r = .870) 1-factor 2-factor (r = .874) 1-factor 2-factor (r = .878)
AFF SOM AFF SOM AFF SOM AFF SOM
Item 1 .84 .86 .79 .81 .84 .86 .82 .84
Item 2 .79 .81 .81 .84 .76 .78 .81 .83
Item 3 .73 .74 .74 .76 .71 .74 .74 .76
Item 4 .75 .78 .77 .79 .77 .80 .75 .77
Item 5 .73 .75 .80 .83 .73 .76 .75 .78
Item 6 .73 .82 .77 .80 .84 .86 .77 .79
Item 7 .77 .80 .78 .80 .77 .81 .77 .80
Item 8 .76 .78 .74 .77 .80 .82 .74 .76
Open in a new tab

Notes: AFF = Affective, SOM = Somatic. Values inside the parenthesis are the correlations between latent variables for 2-factor model.

Hypothesis 2 examined whether the factor structure was invariant for (1) Mexicans and Central Americans and (2) men and women (see Table 4). For Mexican and Central American descent university students, the configural model (i.e., all parameters were free) had acceptable fit (χ2(40) = 730.371, p < .05, CFI = .969). Next, constrained factor loadings did not worsen the fit outside the set criteria (i.e., χ2(47) = 702.232, p < .05, CFI = .970, ΔCFI = −.001; Δχ2(7) = 9.368, p > .05). Hence, the PHQ-8 met metric invariance. Afterward, the loadings and thresholds (i.e., scalar invariance) were constrained and met criteria (i.e., χ2(62) = 608.708, p < .05, CFI = .975, ΔCFI = − .0005; Δχ2(15) = 16.641, p > .05). Hence, the PHQ-8 met scalar invariance between Mexican and Central American descent university students given the joint criteria.

Table 4.

PHQ-8’s Group Equivalence Test

Model χ2 ΔTest CFI ΔCFI
χ2 df p χ2 df p
Mexican vs. Central American descent
 Configural(C) 730.37 40 <.001 .969
 Metric(M) 702.23 47 <.001 .970
  M vs C 9.368 7 .227 .001
 Scalar(S) 608.71 62 <.001 .975 .005
  S vs M 16.641 15 .341
Men vs Women
 Configural 722.040 40 <.001 .969
 Metric 751.332 47 <.001 .968 .001
  M vs C 53.284 7 <.001
 Scalar 724.059 62 <.001 .969 .003
  S vs M 73.322 15 <.001
Open in a new tab

Note. CFI were reported to the three decimals.

For male and female university students, the configural model (i.e., all parameters were free) met fit criteria (χ2(40) = 722.040, p < .05, CFI = .969). Next, constrained factor loadings did not worsen fit outside the set criteria (i.e., χ2(47) = 751.332, p < .05, CFI = .968, ΔCFI = −.001; Δχ2(7) = 53.284, p < .05). Next, the thresholds (i.e., scalar invariance) were constrained and met criteria to support invariance (i.e., χ2(62) = 724.059, p < .05, CFI = .969 ΔCFI = −.001; Δχ2(15) = 73.322, p < .05). Recall we used a dual criteria to judge a lack of invariance which included the combination of a significant Δχ2 and ΔCFI < .01. To address the reviewer’s concern that the analyses confound culture with sex, analyses were conducted with all possible sex and cultures comparisons (e.g., Mexican American women vs. Central American men; Mexican American men vs. Central American women). We did not report these results because the conclusions were the same as presented; that is, models in all comparisons reached full invariance.

Validity Evidence with Associations with Other Variables

Hypothesis 3 examined associations between the PHQ-8 and other variables for Mexican and Central American, as well as Latina/o men and women. Table 5 summarizes the findings of the association between the PHQ-8 and theoretically related variables (i.e., convergent and concurrent), including the bootstrapping results. In convergent validity, the PHQ-8 was strongly and positively correlated with the previously validated CES-D short version within each group of Latina/o university students (i.e., Mexicans, Central Americans, men, and women). In concurrent validity, the PHQ-8 depression factor was strongly and positively related to perceived stress within each subsample. Also for each subsample, the PHQ-8 depression factor was strongly and negatively related to quality of life and moderately and negatively related to perceived parental support. Hence, the PHQ-8 met convergent and concurrent validity.

Table 5.

Pearson Correlations between PHQ and Theoretically-Related Variables

r Bootstrapping
Bias SE CI-Lower CI-Upper
Mexican descent
 CESD (n = 650) .649** .001 .032 .569 .695
 Quality of life (n = 2032) −.525** .000 .037 −.572 −.425
 Perceived stress (n = 1555) .661** .001 .026 .548 .652
 Parental support (n = 1656) −.236** .001 .045 −.298 −.116
Central American descent
 CESD (n = 239) .693** −.002 .040 .600 .757
 Quality of life (n = 739) −.481** .001 .049 −.607 −.411
 Perceived stress (n = 583) .599** −.001 .054 .411 .625
 Parental support (n = 573) −.269** −.001 .060 −.379 −.138
Men
 CESD (n = 241) .658** .001 .061 .501 .749
 Quality of life (n = 796) −.545** .001 .051 −.692 −.488
 Perceived stress (n = 641) .644** .000 .045 .486 .663
 Parental support (n = 582) −.282** −.003 .072 −.375 −.093
Women
 CESD (n = 239) .663** .001 .029 .597 .710
 Quality of life (n = 794) −.500** .000 .036 −.538 −.394
 Perceived stress (n = 639) .640** .000 .029 .511 .627
 Parental support (n = 590) −.233** −.002 .039 −.302 −.149
Open in a new tab
**

p < .01.

Note: The value in the parenthesis is the sample size for the people that completed those measurements. CESD stands for the Center for Epidemiologic Studies Depression Scale. Also, bootstrapping parameters were estimated, which indicated minimum bias.

Discussion

The purposes of this study were to evaluate whether validity evidence of the PHQ-8 scores would support inferences and actions for (1) Mexican and Central American descent and (2) male and female university students. This population was selected for the following reasons: (1) because of significant life changes and transitions, emerging adult university students are at risk of psychopathology; (2) typically, individuals of Mexican and Central American descent report higher rates of depressive symptoms than other ethnic groups; and (3) there is scarcity of research on evaluating mainstream depression instruments’ psychometric properties for emerging adults from these populations. In addition, there is a need for a quick DSM-5 related screener, given that many of the widely used depression scales are lengthy and deviate from the depression criteria’s symptoms outlined in the DSM-5. From a practical perspective, a short, reliable, and valid depression scale would be most favorable in order to measure depression symptomatology, considering that depression is a major health concern worldwide (Lopez & Mathers, 2006) and in college settings (American College Health Association, 2008; Del Pilar, 2008; Furr, McConnell, Westefeld, & Jenkins, 2001; Voelker, 2003). For these reasons, the current study collected sources of validity evidence to build a validity argument for the inferences and used score for the PHQ-8. Although there are many sources of validity evidence (e.g., internal structure, response process), the current study only focused on internal structure and association with other variables for sources of validity evidence.

In the current study, evidence suggested a 1-factor model might be more appropriate given model fit and high correlations between the two factors. Researchers have documented that the PHQ-8 has a 1-factor structure among Latina/os (Huang et al., 2006; Merz et al., 2011), and also among men of Mexican descent (Donlan & Lee, 2010). The current study’s findings support the PHQ-8’s unidimensional structure for Mexican descent, Central American descent, men, and women university students. This indicated that item responses of the PHQ-8 are driven by the depression factor. Having one dimension for this scale is important because most researchers and practitioners are likely to use the PHQ-8 composite score (i.e., average or sum of all items) for the quantification of depressive symptoms.

In addition, results supported the invariance of the scale between (1) Mexican and Central American descent and (2) men and women. This indicated that the PHQ-8 has the potential to be used because it captures the same construct between these groups and within these groups. In the current study, metric and scalar invariance was demonstrated between (1) Mexican and Central American descent and (2) men and women. In the future, researchers and theorists should further investigate the factor structure of the PHQ-8 across different ethnic groups and evaluate the factor invariance because the PHQ-8 lacks research across many ethnic groups and may not be adequately capturing depressive symptoms for these groups.

Next, we examined the association of the PHQ-8 with theoretically related variables. Initially, the convergent validity of the PHQ-8 was examined with an established measure of depression (i.e., CES-D). In the current study, the PHQ-8 total score was strongly related to the 10-item version of the CES-D in each of the subsamples. The current findings support previous research showing that the PHQ-8 and PHQ-9 are highly correlated with well-established depression scales (e.g., Ganguly et al., 2013; Kroenke et al., 2010; Kroenke et al., 2009). Our results are important as they demonstrated that the PHQ-8 captures the same construct as other established depression inventories. The results provided evidence that the scores on the PHQ-8 are associated with established measures at the strength and direction hypothesized. Thus, this strengthens the validity argument.

Also, the concurrent validity of the PHQ-8 was examined for Mexicans, Central Americans, men, and women. The PHQ-8 total score was significantly correlated with theoretically related variables in the expected directions across the subsamples. In this regard, the PHQ-8 was positively associated to perceived stress, and negatively correlated to perceived parental support and quality of life. The current findings supported the results of prior studies that (1) stress can exacerbate depressive symptoms (e.g., Flores et al., 2008; Ng & Jeffery, 2003; Vega et al., 2010); (2) perceived parental support is negatively related to depressive symptoms (e.g., Liem et al., 2010; Needham, 2008); and (3) depressive symptoms are detrimental to quality of life (e.g., Abraham & Fava, 1999; Gotlib, 1997). Overall, the PHQ-8 was related with the theoretically related variables in the hypothesized directions among the target populations. Hence, the current study established the concurrent validity of the PHQ-8. Also, since the PHQ-8 was related to theoretically related variables in expected directions, the PHQ-8 should be a good indicator of depressive symptomatology for researchers studying predictors of depression as well as outcomes of depression among Mexican and Central American descent, men, and women university students and emerging adults.

Limitations and Research Implications

This study contributed to documenting the psychometric properties of this wide-used scale for use with young adults from (1) Mexican and Central American descent and (2) men and women. However, there are certain limitations of the present investigation that need to be addressed. First, the participants were recruited from psychology classes at a Southern California university. Thus, this limits the generalizability of the present findings. In future studies, researchers should examine the psychometric properties of the PHQ-8 in samples of non-college emerging adults of Mexican and Central American descent and other Latina/o emerging adults, as well as in other ethnic minority samples. Second, the current study did not take into consideration differences by generational status or years in the United States for immigrants. Third, the Central American sample was collapsed into one group, which is not accounting for differences in the different nationality and geographical region between these groups. We were not able to separate out groups due to sample size of a given area. In the future, investigators should examine intragroup differences (e.g., generation statuses, country of origin). Fourth, most of the sample (i.e., approximately two thirds) answered the protocol online while the remaining data were collected in groups using paper-and-pencil questionnaires. Future studies should evaluate the format since it can play a role in self-report survey. Finally, the current study did not collect participants’ prior history of depression, highly co-morbid illnesses (e.g., anxiety disorders or substance addictions), or family history of depression. This individual differences can potentially play a role. Researchers should evaluate the PHQ-8’s psychometric properties with individuals who have a personal and family history of depression or co-morbid illness. Overall, generalizing results was limited, yet the current results showed that the PHQ-8 seemed to be a valid and practical self-report measure of depression for male and female college students of Mexican and Central American descent.

Implication and Conclusion

In summary, the PHQ-8 produced similar results for Mexican and Central American descent, men, and women university students. Also, evidence was provided for a unidimensional depression inventory and for concurrent and convergent validity of the PHQ-8. Community mental health care practitioners, college counselors, medical professionals, as well as social scientists can utilize the PHQ-8 with the target populations.

What is the public health significance of this article?

This study found the 8-item Patient Health Questionnaire (PHQ-8) was a valid scale to measure levels of depression among male and female young adults of Mexican and Central American descent in Los Angeles. Thus, mental health practitioners, college counselors, medical professionals, and researchers could use this free, short, and valid depression scale in their work with these young adults in the United States. This is significant as depression has serious health consequences and influences the lives of millions of individuals, especially Latina/o college students who are a high-risk group for depression.

Contributor Information

David Alpizar, Email: d.martinezalpizar@wsu.edu, Educational Psychology, Washington State University, WA 99164-2136

Luciana Lagana, Email: luciana.lagana@csun.edu, Professor, Department of Psychology, California State University Northridge, Northridge, CA 91330-8255

Scott W. Plunkett, Email: scott.plunkett@csun.edu, Professor, Department of Psychology, California State University Northridge, Northridge, CA 91330-8255

Brian F. French, Email: frenchb@wsu.edu, Professor, Educational Psychology, Washington State University, WA 99164-2136

References

  1. Abraham HD, Fava M. Order of onset of substance abuse and depression in a sample of depressed outpatients. Comprehensive Psychiatry. 1999;40(1):44–50. doi: 10.1016/S0010-440X(99)90076-7. [DOI] [PubMed] [Google Scholar]
  2. Adkins DE, Wang V, Dupre ME, van den Oord EJCG, Elder GH. Structure and stress: Trajectories of depressive symptoms across adolescence and young adulthood. Social Forces. 2009;88(1):31–60. doi: 10.1353/sof.0.0238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. American College Health Association. American College Health Association -National college health assessment spring 2007 reference group data report (abridged) Journal of American College Health. 2008;56(5):469–479. doi: 10.3200/jach.56.5.469-480. [DOI] [PubMed] [Google Scholar]
  4. American Heart Association. Anxiety, depression identify heart disease patients at increased risk of dying. Dallas, TX: Author; 2013. Mar 19, Retrieved March 11, 2014, from http://newsroom.heart.org/news/anxiety-depression-identify-heart-disease-patients-at-increased-risk-of-dying. [Google Scholar]
  5. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5. Washington, DC: Author; 2013. [Google Scholar]
  6. Arnett JJ. Emerging adulthood: The cultural psychology of a new life stage. In: Jensen LA, editor. Bridging cultural and developmental approaches to psychology: New syntheses in theory, research and policy. New York, NY: Oxford University Press; 2011. pp. 255–275. [Google Scholar]
  7. Banks JA. Handbook of research on multicultural education. New York, NY: Macmillan; 1995. [Google Scholar]
  8. Bishara AJ, Hittner JB. Reducing bias and error in the correlation coefficient due to nonnormality. Educational and Psychological Measurement. 2014;75(5):785–804. doi: 10.1177/0013164414557639. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Behnke AO, Plunkett SW, Sands T, Bámaca-Colbert MY. The relationship between Latino adolescents’ perceptions of discrimination, neighborhood risk, and parenting on self-esteem and depressive symptoms. Journal of Cross-Cultural Psychology. 2011;42(7):1179–1197. doi: 10.1177/0022022110383424. [DOI] [Google Scholar]
  10. Brown JS, Meadows SO, Elder GH., Jr Race-ethnic inequality and psychological distress: Depressive symptoms from adolescence to young adulthood. Developmental Psychology. 2007;43(6):1295–1311. doi: 10.1037/0012-1649.43.6.1295. [DOI] [PubMed] [Google Scholar]
  11. Bush KR, Peterson GW, Cobas JA, Supple AJ. Adolescents’ perceptions of parental behaviors as predictors of adolescent self-esteem in mainland China. Sociological Inquiry. 2002;72:503–526. doi: 10.1111/1475-682X.00031. [DOI] [Google Scholar]
  12. Camacho Á, Gonzalez P, Buelna C, Emory K, Talavera G, Castañeda S, … Isasi CR. Anxious-depression among Hispanic/Latinos from different backgrounds: Results from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) Social Psychiatry & Psychiatric Epidemiology. 2015;50(11):1669–1677. doi: 10.1007/s00127-015-1120-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen FF. What happens if we compare chopsticks with forks? The impact of making inappropriate comparisons in cross-cultural research. Journal of Personality and Social Psychology. 2008;95(5):1005–1018. doi: 10.1037/a0013193. [DOI] [PubMed] [Google Scholar]
  14. Chen FF. Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling. 2007;14(3):464–504. doi: 10.1080/10705510701301834. [DOI] [Google Scholar]
  15. Cohen S, Williamson G. Perceived stress in a probability sample of the United States. In: Spacapan S, Oskamp S, editors. The social psychology of health: Claremont symposium on applied social psychology. Newbury Park, CA: Sage; 1988. pp. 31–67. [Google Scholar]
  16. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. Journal of Health and Social Behavior. 1983;24(4):385–396. doi: 10.2307/2136404. [DOI] [PubMed] [Google Scholar]
  17. Crockett LJ, Randall BA, Shen Y, Russell ST, Driscoll AK. Measurement equivalence of the Center for Epidemiological Studies Depression Scale for Latino and Anglo adolescents: A national study. Journal of Consulting and Clinical Psychology. 2005;73(1):47–58. doi: 10.1037/0022-006X.73.1.47. [DOI] [PubMed] [Google Scholar]
  18. Del Pilar JA. Mental health and Latino/a college students: A psychological perspective and new findings. Journal of Hispanic Higher Education. 2008;8:263–281. doi: 10.1177/1538192708328891. [DOI] [Google Scholar]
  19. Donlan W, Lee J. Screening for depression among indigenous Mexican Immigrant farm workers using the Patient Health Questionnnaire-9. Psychological Reports. 2010;106(2):419–432. doi: 10.2466/pr0.106.2.419-432. [DOI] [PubMed] [Google Scholar]
  20. Eisenberg D, Gollust SE, Golberstein E, Hefner JL. Prevalence and correlates of depression, anxiety, and suicidality among university students. American Journal of Orthopsychiatry. 2007;77(4):534–542. doi: 10.1037/0002-9432.77.4.534. [DOI] [PubMed] [Google Scholar]
  21. Fazel M, Wheeler J, Danesh J. Prevalence of serious mental disorder in 7000 refugees resettled in western countries: a systematic review. The Lancet. 2005;365(9467):1309–1314. doi: 10.1016/s0140-6736(05)61027-6. [DOI] [PubMed] [Google Scholar]
  22. Field A. Discovering statistics using IBM SPSS statistics. Thousand Oaks, CA: Sage; 2013. [Google Scholar]
  23. Finney SJ, DiStefano C. Nonnormal and categorical data in structural equation models. In: Hancock GR, Mueller RO, editors. A second course in structural equation modeling. 2. Charlotte, NC: Information Age; 2013. pp. 439–492. [Google Scholar]
  24. Flores E, Tschann JM, Dimas JM, Bachen EA, Pasch LA, de Groat CL. Perceived discrimination, perceived stress, and mental and physical health among Mexican-origin adults. Hispanic Journal of Behavioral Sciences. 2008;30(4):401–424. doi: 10.1177/0739986308323056. [DOI] [Google Scholar]
  25. Cheung GW, Rensvold RB. Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling. 2002;9(2):233–255. doi: 10.1207/s15328007sem0902_5. [DOI] [Google Scholar]
  26. Cohen J. Statistical power analysis for the behavioral sciences. 2. Hillsdale, NJ: Erlbaum; 1988. [Google Scholar]
  27. Familiar I, Ortiz-Panozo E, Hall B, Vieitez I, Romieu I, Lopez-Ridaura R, Lajous M. Factor structure of the Spanish version of the Patient Health Questionnaire-9 in Mexican women. International Journal of Methods in Psychiatric Research. 2014;24(1):74–82. doi: 10.1002/mpr.1461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Furr SR, McConnell GN, Westefeld JS, Jenkins MJ. Suicide and depression among college students: A decade later. Professional Psychology. 2001;32:97–100. doi: 10.1037/0735-7028.32.1.97. [DOI] [Google Scholar]
  29. Ganguly S, Samanta M, Roy P, Chatterjee S, Kaplan DW, Basu B. Patient Health Questionnaire-9 as an effective tool for screening of depression among Indian adolescents. Journal of Adolescent Health. 2013;52(5):546–551. doi: 10.1016/j.jadohealth.2012.09.012. [DOI] [PubMed] [Google Scholar]
  30. Gay LR. Educational research. Upper Saddle River, NJ: Prentice-Hall; 1996. [Google Scholar]
  31. Gonzalez O, Berry JT, McKnight-Eily LR, Strine T, Edwards VJ, Lu H, Croft JB. Current depression among adults – United States, 2006 and 2008. MMWR. 2010;59(38):1229–1235. doi: 10.1037/e664792010-002. [DOI] [PubMed] [Google Scholar]
  32. Gotlib I. The cognitive psychology of depression. East Sussex, UK: Psychology Press; 1997. [Google Scholar]
  33. Granillo M. Structure and function of the Patient Health Questionnaire-9 among Latina and non-Latina White female college students. Journal of the Society for Social Work and Research. 2012;3(2):80–93. doi: 10.5243/jsswr.2012.6. [DOI] [Google Scholar]
  34. Grzywacz JG, Hovey JD, Seligman LD, Arcury TA, Quandt SA. Evaluating short-form versions of the CES-D for measuring depressive symptoms among immigrants from Mexico. Hispanic Journal of Behavioral Sciences. 2006;28:404–424. doi: 10.1177/0739986306290645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Hill PL, Jackson JJ, Roberts BW, Lapsley DK, Brandenberger JW. Change you can believe in: Changes in goal-setting during emerging and young adulthood predict later adult well-being. Social Psychological and Personality Science. 2011;2:123–131. doi: 10.1177/1948550610384510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Hu LT, Bentler PM. Fit indices in covariance structure modeling: Sensitivity to under parameterized model misspecification. Psychological Methods. 1998;3(4):424–453. doi: 10.1037/1082-989x.3.4.424. [DOI] [Google Scholar]
  37. Huang FY, Chung H, Kroenke K, Delucchi KL, Spitzer RL. Using the Patient Health Questionnaire-9 to measure depression among racially and ethnically diverse primary care patients. Journal of General Internal Medicine. 2006;21(6):547–552. doi: 10.1111/j.1525-1497.2006.00409.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Husson O, Mols F, Van de Poll-Franse LV. The relation between information provision and health-related quality of life, anxiety and depression among cancer survivors: A systematic review. Annals of Oncology. 2010;22(4):761–772. doi: 10.1093/annonc/mdq413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Jetelina KK, Reingle Gonzalez JM, Vaeth PC, Mills BA, Caetano R. An investigation of the relationship between alcohol use and major depressive disorder across Hispanic national groups. Alcoholism: Clinical & Experimental Research. 2016;40(3):536–542. doi: 10.1111/acer.12979. [DOI] [PubMed] [Google Scholar]
  40. Kane MT. Validating the interpretations and uses of test scores. Journal of Educational Measurement. 2013;50(1):1–73. doi: 10.1111/jedm.12000. [DOI] [Google Scholar]
  41. Kline RB. Principles and practice of structural equation modeling. New York, NY: Guilford Press; 2005. [Google Scholar]
  42. Kohout FJ, Berkman LF, Evans DA, Cornoni-Huntley J. Two shorter forms of the CES-D depression symptoms index. Journal of Aging and Health. 1993;5:179–193. doi: 10.1177/089826439300500202. [DOI] [PubMed] [Google Scholar]
  43. Kroenke K, Spitzer RL. The PHQ-9: A new depression diagnostic and severity measure. Psychiatric Annals. 2002;32(9):1–7. doi: 10.3928/0048-5713-20020901-06. [DOI] [Google Scholar]
  44. Kroenke K, Spitzer RL, Williams JB, Löwe B. The patient health questionnaire somatic, anxiety, and depressive symptom scales: A systematic review. General Hospital Psychiatry. 2010;32(4):345–359. doi: 10.1016/j.genhosppsych.2010.03.006. [DOI] [PubMed] [Google Scholar]
  45. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The PHQ-8 as a measure of current depression in the general population. Journal of Affective Disorders. 2009;114(1–3):163–173. doi: 10.1016/j.jad.2008.06.026. [DOI] [PubMed] [Google Scholar]
  46. Lehmann LP, Hubbard JR, Martin PR. Substance abuse and depression. In: Hubbard JR, Martin PR, editors. Substance abuse in the mentally and physically disabled. New York, NY: Marcel Dekker; 2001. pp. 33–58. [Google Scholar]
  47. Liem JH, Lustig K, Dillon C. Depressive symptoms and life satisfaction among emerging adults: A comparison of high school dropouts and graduates. Journal of Adult Development. 2010;17(1):33–43. doi: 10.1007/s10804-009-9076-9. [DOI] [Google Scholar]
  48. Lopez AD, Mathers CD. Measuring the global burden of disease and epidemiological transitions: 2002–2030. Annals of Tropical Medicine and Parasitology. 2006;100(5–6):481–499. doi: 10.1179/136485906X97417. [DOI] [PubMed] [Google Scholar]
  49. Merz EL, Malcarne VL, Roesch SC, Riley N, Sadler G. A multigroup confirmatory factor analysis of the Patient Health Questionnaire-9 among English- and Spanish-speaking Latinas. Cultural Diversity and Ethnic Minority Psychology. 2011;17(3):309–316. doi: 10.1037/a0023883. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Moriarty AS, Gilbody S, McMillan D, Manea L. Screening and case finding for major depressive disorder using the Patient Health Questionnaire (PHQ-9): A meta-analysis. General Hospital Psychiatry. 2015;37(6):567–576. doi: 10.1016/j.genhosppsych.2015.06.012. [DOI] [PubMed] [Google Scholar]
  51. Needham BL. Reciprocal relationships between symptoms of depression and parental support during the transition from adolescence to young adulthood. Journal of Youth and Adolescence. 2008;37(8):893–905. doi: 10.1007/s10964-007-9181-7. [DOI] [Google Scholar]
  52. Ng DM, Jeffery RW. Relationships between perceived stress and health behaviors in a sample of working adults. Health Psychology. 2003;22(6):638–642. doi: 10.1037/0278-6133.22.6.638. [DOI] [PubMed] [Google Scholar]
  53. Nguyen HT, Kitner-Triolo M, Evans MK, Zonderman AB. Factorial invariance of the CES-D in low socioeconomic status African Americans compared with a nationally representative sample. Psychiatry Research. 2004;126(2):177–187. doi: 10.1016/j.psychres.2004.02.004. [DOI] [PubMed] [Google Scholar]
  54. Oquendo MA, Lizardi D, Greenwald S, Weissman MM, Mann JJ. Rates of lifetime suicide attempt and rates of lifetime major depression in different ethnic groups in the United States. Acta Psychiatrica Scandinavica. 2004;110(6):446–451. doi: 10.1111/j.1600-0447.2004.00404.x. [DOI] [PubMed] [Google Scholar]
  55. Peterson GW. Family influences on adolescent development. In: Gullotta TP, Adams GR, editors. Handbook of adolescent behavioral problems: Evidence-based approaches to prevention and treatment. New York, NY: Springer; 2005. pp. 27–55. [Google Scholar]
  56. Radloff L. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;7:385–401. doi: 10.1177/014662167700100306. [DOI] [Google Scholar]
  57. Razykov I, Ziegelstein RC, Whooley MA, Thombs BD. The PHQ-9 versus the PHQ-8—Is item 9 useful for assessing suicide risk in coronary artery disease patients? Data from the Heart and Soul Study. Journal of Psychosomatic Research. 2012;73(3):163–168. doi: 10.1016/j.jpsychores.2012.06.001. [DOI] [PubMed] [Google Scholar]
  58. Roberti JW, Harrington LN, Storch EA. Further psychometric support for the 10-item version of the Perceived Stress Scale. Journal of College Counseling. 2006;9(2):135–147. doi: 10.1002/j.2161-1882.2006.tb00100.x. [DOI] [Google Scholar]
  59. Schwarz N. Self-reports: How the questions shape the answers. American Psychologist. 1999;54(2):93–105. doi: 10.1037/0003-066x.54.2.93. [DOI] [Google Scholar]
  60. Spitzer R, Williams J, Kroenker K, Linzer M, Degruy F, Hahn S, Brody D, Johnson J. Utility of a new procedure for diagnosing mental disorders in primary care. The PRIME-MD 1000 study. JAMA. 1994;272(22):1749–1756. doi: 10.1001/jama.1994.03520220043029. [DOI] [PubMed] [Google Scholar]
  61. Spitzer RL, Kroenke K, Williams JB Patient Health Questionnaire Primary Care Study Group. Validation and utility of a self-report version of PRIME-MD: The PHQ primary care study. JAMA. 1999;282(18):1737–1744. doi: 10.1001/jama.282.18.1737. [DOI] [PubMed] [Google Scholar]
  62. Tabachnick GB, Fidell SL. Using multivariate statistics. Upper Saddle River, NJ: Pearson Education; 2012. [Google Scholar]
  63. Thompson B, Daniel LG. Factor analytic evidence for the construct validity of scores: A historical overview and some guidelines. Educational and Psychological Measurement. 1996;56(2):197–208. doi: 10.1177/0013164496056002001. [DOI] [Google Scholar]
  64. Umaña-Taylor AJ, Guimond AB. A longitudinal examination of parenting behaviors and perceived discrimination predicting Latino adolescents’ ethnic identity. Developmental Psychology. 2010;46(3):636–650. doi: 10.1037/a0019376. [DOI] [PubMed] [Google Scholar]
  65. U.S. Preventive Services Task Force. Screening for depression: Recommendations and rationale. Annals of Internal Medicine. 2002;136(10):760–764. doi: 10.7326/0003-4819-136-10-200205210-00012. [DOI] [PubMed] [Google Scholar]
  66. Vega WA, Rodriguez MA, Ang A. Addressing stigma of depression in Latin primary care patients. General Hospital Psychiatry. 2010;32(2):181–191. doi: 10.1016/j.genhosppsych.2009.10.008. [DOI] [PubMed] [Google Scholar]
  67. Voelker R. Mounting student depression taxing campus mental health services. JAMA. 2003;289(16):2055–2056. doi: 10.1001/jama.289.16.2055. [DOI] [PubMed] [Google Scholar]
  68. Zambrana RE, Scrimshaw SM, Collins N, Dunkel-Schetter C. Prenatal health behaviors and psychosocial risk factors in pregnant women of Mexican origin: The role of acculturation. American Journal of Public Health. 1997;87(6):1022–1026. doi: 10.2105/AJPH.87.6.1022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Zimmerman M, Ruggero CJ, Chelminski I, Young D, Posternak MA, Friedman M, … Attiullah N. Developing brief scales for use in clinical practice: The reliability and validity of single-item self-report measures of depression symptom severity, psychosocial impairment due to depression, and quality of life. Journal of Clinical Psychiatry. 2006;67(10):1536–1541. doi: 10.4088/JCP.v67n1007. [DOI] [PubMed] [Google Scholar]

RESOURCES