Concurrent and Prospective Prediction of Community-dwelling Adults’ Psychosocial Functioning with the Inventory of Depression and Anxiety Symptoms-II (IDAS-II)

Abstract

Mood and anxiety disorders involve defining symptoms (e.g., dysphoria, anhedonia) that can impair psychosocial functioning (e.g., self-care, work, social relationships). The current study evaluated the validity of the Inventory of Depression and Anxiety Symptoms-II (IDAS-II; Watson et al., 2012) via convergence with a semi-structured interview assessing mood and anxiety disorder symptoms and, moreover, prediction of psychosocial functioning. Community-dwelling adults (N = 601) completed the self-report IDAS-II, a semi-structured diagnostic interview, and self-report and interview measures of psychosocial functioning. A retest subsample (ns 497–501) completed the functioning measures again, on average 8 months later. Supporting our hypotheses, the IDAS-II converged robustly with interview-assessed symptoms and predicted psychosocial functioning significantly, both concurrently and prospectively. Moreover, the IDAS-II predicted functioning significantly better than did the diagnostic interview. These findings support use of the IDAS-II in research and clinical settings to assess mood and anxiety symptoms and their connections to psychosocial impairment.

Public Significance Statement

Anxiety and mood disorders involve key symptoms, such as emotional distress and lack of interest or pleasure, that interfere with functioning in areas including self-care, work, and social relationships. We found that the Inventory of Depressive and Anxiety Symptoms-II (IDAS-II) is a valid measure of key symptoms and a good predictor of functioning among community-dwelling adults, so the IDAS-II may be useful for both research and patient care.

Mood and anxiety disorders have been defined traditionally as heterogeneous sets of symptoms that impair psychological functioning (American Psychiatric Association, 2013; World Health Organization, 2022). For example, major depressive disorder could include dysphoria, weight loss, agitation, insomnia, and fatigue or, for the same diagnosis, anhedonia, weight gain, psychomotor slowing, hypersomnia, and suicidality, as long as the count reaches at least five from a list of nine symptoms (American Psychiatric Association, 2013). This polythetic diagnostic approach differs from psychometrically driven efforts to differentiate multiple homogeneous symptom dimensions (e.g., Forbes et al., 2024; Kotov et al., 2022). In this vein, many of the major depressive disorder symptoms listed above can be measured reliably as distinct dimensions (Watson & O’Hara, 2017). The purposes of the current report are to (a) examine convergence between a modern multidimensional self-report measure (Inventory of Depression and Anxiety Symptoms-II (IDAS-II); Watson et al., 2012) and a traditional mood and anxiety diagnostic interview (Mini International Neuropsychiatric Interview (MINI); Sheehan et al., 1997, updated to reflect DSM-5 with the permission of the author) and, moreover, (b) test and compare these assessment approaches’ concurrent and prospective prediction of several domains of psychosocial impairment.

The IDAS-II assesses symptoms of DSM-IV and -5 emotional disorders, including major depressive, generalized anxiety, social anxiety, post-traumatic stress, panic, obsessive-compulsive, and bipolar disorders, as well as phobias (Watson et al., 2007; Watson et al., 2012; Watson & O’Hara, 2017). As listed in Table 1, the IDAS-II’s 19 scales are scored from 99 items, each rated from 1 (not at all) to 5 (extremely) considering the past 2 weeks. Eighteen of the IDAS-II scales are non-overlapping in item content, whereas the general depression scale includes items from the dysphoria, suicidality, lassitude, insomnia, appetite loss, and (reverse-keyed) well-being scales. In contrast to well-being’s coverage of adaptive positive affectivity (e.g., optimism, pleasure), the IDAS-II euphoria (e.g., excess energy, grandiosity) and mania (e.g., pressured speech, flight of ideas) scales assess problematic positive affectivity associated with bipolar disorders. The IDAS-II symptom scales reflect a higher-order internalizing psychopathology spectrum, distinguishable from somatoform, thought disorder, disinhibited externalizing, antagonistic externalizing, and detachment spectra, which can be assessed with other measures (Kotov et al., 2022; Watson et al., 2022). In addition to internalizing psychopathology, the IDAS-II mania and euphoria scales may also partly reflect the thought disorder spectrum.

Table 1.

Descriptive Statistics for Symptom Measures at Assessment Session 1

IDAS-II scales (number of items)	M	SD	Min.	Max.	Alpha
General depression (20)	2.41	0.74	1.00	4.90	.91
Dysphoria (10)	2.37	0.89	1.00	4.80	.89
Lassitude (6)	2.35	0.87	1.00	5.00	.80
Insomnia (6)	2.54	1.01	1.00	5.00	.86
Suicidality (6)	1.33	0.55	1.00	4.33	.78
Appetite loss (3)	2.08	0.98	1.00	5.00	.80
Appetite gain (3)	2.11	1.00	1.00	5.00	.78
Ill temper (5)	1.71	0.85	1.00	5.00	.87
Well-being (8)	2.69	0.89	1.00	5.00	.88
Social anxiety (6)	1.91	0.88	1.00	5.00	.84
Panic (8)	1.72	0.76	1.00	5.00	.85
Claustrophobia (5)	1.63	0.95	1.00	5.00	.88
Cleaning (7)	1.55	0.76	1.00	5.00	.86
Checking (3)	2.00	0.96	1.00	5.00	.78
Ordering (5)	1.89	0.82	1.00	5.00	.78
Mania (5)	1.91	0.90	1.00	5.00	.82
Euphoria (5)	1.61	0.71	1.00	5.00	.77
Traumatic intrusions (4)	1.95	1.02	1.00	5.00	.85
Traumatic avoidance (4)	2.26	1.06	1.00	5.00	.84
MINI diagnostic symptom counts	M	SD	Min.	Max.	ICC
Major depressive disorder	3.50	2.73	0.00	9.00	.99
Dysthymic disorder	2.86	2.15	0.00	6.00	.99
Manic episode	1.18	2.28	0.00	7.00	.91
Panic disorder	2.90	4.35	0.00	13.00	.93
Agoraphobia	0.54	1.52	0.00	5.00	.85
Social anxiety disorder	1.12	1.69	0.00	4.00	.95
Obsessive-compulsive disorder	0.66	1.30	0.00	5.00	.99
Post-traumatic stress disorder	2.84	5.27	0.00	20.00	.99
Generalized anxiety disorder	4.56	3.91	0.00	12.00	.99

Note. Session 1 N = 601. IDAS-II = Inventory of Depressive and Anxiety Symptoms-II. MINI = Mini International Neuropsychiatric Interview. Symptom counts reflected presentation at the time of interview, except for manic episode which was lifetime history. Alpha (internal consistency) and ICC (inter-rater) are reliability estimates.

The IDAS-II scales have shown strong psychometric properties in past research. Watson and O’Hara (2017) provided summaries of reliability and validity data, aggregating findings from several moderate-large samples of adults, patients, and students. Some research utilized the original IDAS (Watson et al., 2007), which contains a subset of 12 of the 19 IDAS-II scales and so applies fully to the IDAS-II. Regarding reliability, the IDAS-II scales’ alpha internal consistency coefficients typically fall in the .75–.90 range. Further, short-term (e.g., 2-week) retest (i.e., dependability; Chmielewski et al., 2009) correlations usually range from about .55–.75 for the IDAS-II scales (Watson & O’Hara, 2017).

Validity evidence for the IDAS-II includes convergence with other mood and anxiety symptom measures and diagnostic interviews (Watson & O’Hara, 2017). The IDAS-II specific symptom scales have converged highly with interviews assessing the same constructs (median correlation = .61, range .37–.77). More directly applicable to the current study, the IDAS-II scales have also converged robustly with categorical anxiety and mood disorder diagnoses derived from the Structured Clinical Interview for DSM-IV (SCID-IV; First et al., 1997) or MINI. For example, as a central marker of the internalizing psychopathology spectrum, the IDAS-II dysphoria scale correlated substantively with several traditional mood and anxiety disorder diagnoses, such as major depressive disorder (.70), generalized anxiety disorder (.45), panic disorder (.41), post-traumatic stress disorder (.40), agoraphobia (.30), and social phobia (.28). More diagnosis-specific convergence with IDAS-II scales included major depressive disorder’s correlations with lassitude (.51), suicidality (.44), insomnia (.43), and appetite loss (.36); manic episode with mania (.40) and euphoria (.34); post-traumatic stress disorder with traumatic intrusions (.58) and traumatic avoidance (.55); social phobia with social anxiety (.52); panic disorder with panic (.48); agoraphobia with claustrophobia (.47); specific phobia with claustrophobia (.42); and obsessive-compulsive disorder with ordering (.50), checking (.46), and cleaning (.40). Thus, the IDAS-II scales can differentiate common and specific symptoms of traditionally-defined mood and anxiety disorders.

Although past research has been limited, the IDAS-II has shown concurrent correlations with psychosocial functioning measures. For example, among Latina mothers living in North Carolina (Santos et al., 2021), the IDAS-II correlated with self-report measures of discrimination, acculturative stress, and economic hardship. At 4–6 weeks postpartum, the mean absolute correlations with the IDAS-II scales overall (and with the two strongest correlating IDAS-II scales, respectively) were .26 for everyday discrimination (general depression = .39, checking = .36), .20 for acculturation stress (checking = .29, mania = .26), and .12 for economic hardship (traumatic intrusions = .27, dysphoria = .21). Similarly, among Spanish adults (De la Rosa-Cáceres et al., 2023) who completed a translation of the IDAS-II (De la Rosa-Cáceres et al., 2020), the IDAS-II scales collectively accounted for 57% of variance in a global index of psychological impairment in a multiple regression analysis. The IDAS-II scales showing the strongest unique relations (standardized betas) with the psychosocial impairment measure included general depression (.24), panic (.17), claustrophobia (.15), and lassitude (.14).

The IDAS-II is fully multidimensional, in the sense that each specific symptom scale indexes a quantitative severity continuum or dimension, each continuum is homogeneous (e.g., its items all reflect the same underlying construct), and the continua are distinct from one another (i.e., have demonstrated discriminant validity with respect to external variables; Hopwood et al., 2023; Watson & O’Hara, 2017). Other dimensional measures may have fewer of these desirable properties and thus provide less useful information. For example, diagnostic symptom counts are continuous but often heterogeneous/polythetic and less distinct from one another (Hopwood et al., 2023). When the hundreds of DSM-5 symptoms are allowed to covary freely in analyses, they clearly do not re-aggregate into their nominal diagnostic categories but instead form factors similar to the symptom dimensions and internalizing spectrum captured by the IDAS-II (Forbes et al., 2024). Therefore, whether the IDAS-II’s structural validity strengthens its prediction of psychosocial functioning relative to diagnostic symptom counts was evaluated in the present study.

To understand the validity of the IDAS-II more fully, we tested three hypotheses in a sample of community-dwelling adults who completed two assessment sessions, on average 8 months apart. First, we hypothesized that the IDAS-II self-report would converge with mood and anxiety disorder symptom counts from the MINI at session 1. We hypothesized, second and third, that the IDAS-II would predict psychosocial functioning concurrently (session 1) and prospectively (session 1 symptoms, session 2 functioning). Psychosocial functioning was assessed both by interview and self-report. Finally, we compared the IDAS-II and MINI’s predictions of psychosocial functioning but did not make specific hypotheses.

By testing these associations, the current analyses made several contributions to the literature. We expanded the research base regarding convergence of the IDAS-II with a traditional diagnostic assessment (MINI) in a new, relatively large sample of adults. Moreover, we provided finer-grained data on the IDAS-II’s relations to multiple dimensions of psychosocial functioning (e.g., work, leisure, relationships), concurrently and prospectively, assessed by self-report and interview, alone and in comparison with the MINI. Psychosocial impairment is integral to definitions of disorder (American Psychiatric Association, 2013; World Health Organization, 2022), and therefore also central in evaluating the validity and utility of symptom assessments. Of note, we analyzed diagnostic symptom counts, instead of the categorical diagnoses derived from those counts, because imposing cut-points can decrease validity (Markon et al., 2011) and internalizing symptoms relate linearly to impairment throughout their severity ranges (Markon, 2010). Thus, our comparisons of the IDAS-II and MINI’s predictions of functioning were arguably more conservative (i.e., not biased in favor of continuous scales) than analyses utilizing categorical diagnoses.

Method

Transparency and Openness

The study design and analyses reported in this article were not preregistered. The measures utilized in this article have not been posted in a repository but may be available from the measures’ authors or publishers. For example, the IDAS-II is freely available for non-commercial clinical and research use from its authors. The data presented in this article are available upon request, for the purpose of verifying that the results reported are veridical. Please address data requests to Dr. Lee Anna Clark. Data analysis R scripts are available on request from Dr. Jeffrey Vittengl.

Participants

Participants (N = 601) were community-dwelling adults, at least 18 years old, able to respond to interviews and questionnaires in English, and without dementia, intellectual disability, or active psychosis. Participants were M = 46 (SD = 13) years old; 57% were women; 68% were white, 21% black, and 11% other racial/ethnic groups. Participants included mental health outpatients (50%) and adults at high risk for personality pathology but not in mental health treatment (50%), recruited as described below. The Institutional Review Board of the University of Notre Dame approved the study procedures, and all participants provided informed consent.

Outpatients were referred by a community mental-health center and local practitioners. When current mental-health outpatient status could be verified, a few outpatients were also recruited by the methods used to recruit high-risk adults: Newsletters, listservs, and mass emails sent to University of Notre Dame staff, faculty and graduate students, and by word of mouth. Participants at high-risk for personality pathology were contacted via random dialing of landlines and cell telephone numbers in the greater South Bend, Indiana, area. High-risk participants were not receiving treatment from a mental health provider when recruited but gave 2 or more positive responses on the Iowa Personality Disorder Screen, a cut-point with good sensitivity and specificity in identifying personality disorder (Langbehn et al., 1999). A few high-risk participants from the study’s pilot phase (Clark & Ro, 2014) were also recruited and re-screened for eligibility.

Procedure

Participants gave written informed consent and completed measures at a research facility. Participants completed measures over 1–2 days, with breaks for food and beverages to maintain focus and reduce fatigue. Participants answered questionnaires alone on a computer or, in rare exceptions (e.g., due to unfamiliarity with computers or limited eyesight), with the assistance of research staff. Trained interviewers (research study staff, graduate students, and Ph.D.-level researchers) administered structured interviews. Participants agreed to return for additional assessment approximately 6 months later. Researchers used multiple methods to attempt to contact all participants for reassessment, completed on average 8 months later. The IDAS-II, MINI, Longitudinal Interval Follow-up Evaluation: Range of Impaired Functioning Tool (LIFE-RIFT), and World Health Organization Disability Assessment Schedule 2.0 (WHODAS-2) were administered at both session 1 and 2. The current analyses utilized data from the IDAS-II and MINI at session 1, and from the LIFE-RIFT and WHODAS-2 at sessions 1 and 2.

Measures

The IDAS-II is a multidimensional questionnaire consisting of 99 items rated on a 5-point scale of symptom intensity over the past 2 weeks (Watson et al., 2012). The 19 IDAS-II scales are listed in Table 1. The IDAS-II has strong psychometric properties (i.e., internal consistency, convergent and discriminant validity) and demonstrated substantial criterion and incremental validity in relation to interview-based measures of DSM-IV symptoms and disorders (Stasik-O’Brien et al., 2019; Watson et al., 2007, 2012). Internal consistency reliability was high for the IDAS-II scales in the current sample (see Table 1).

The MINI assesses symptoms of 15 common DSM-IV Axis I and ICD-10 psychiatric disorders and has shown strong convergence with other diagnostic interviews (Sheehan et al., 1997; Sheehan et al., 1998). For the current study, the MINI was adapted with permission of its author to assess DSM-5 symptoms (see Table 1). Because data collection for the current study began before publication of DSM-5, one disorder’s assessment utilized proposed criteria that were mostly but not fully aligned with the final published DSM-5 (i.e., generalized anxiety disorder was assessed considering the past 3 instead of 6 months, and avoidance behaviors but not difficulty controlling worry were queried directly). We analyzed interview data regarding current symptom counts, with the exception of lifetime history of manic episode symptoms. Inter-rater reliability estimates (intraclass correlations; Shrout & Fleiss, 1979, formula [1,1]) were high in the current sample (see Table 1), based on data from 13 interviewers who re-rated 54 interview audio recordings.

The LIFE-RIFT (Leon et al., 1999) measures psychosocial dysfunction in three domains, work (paid employment, housework, and schoolwork), social relationships (partner, children, extended family, and friends), and leisure/recreational activities over the past month. Rated by interviewers on a scale from 1 (very good) to 5 (very poor), we analyzed the maximum score in each domain, as well as participants’ self-reported global satisfaction with their functioning. Consistent with past research (Leon et al., 1999), we found high inter-rater reliability for LIFE-RIFT ratings (see Table 2).

Table 2.

Descriptive Statistics for Psychosocial Functioning Measures

	|---------------------Session 1---------------------|					|---------------------Session 2---------------------|
LIFE-RIFT ratings	M	SD	Min.	Max.	ICC	M	SD	Min.	Max.	Retest r
Work dysfunction	2.40	0.90	1.00	5.00	.92	2.36	0.90	1.00	5.00	.56
Relationship dysfunction	2.53	0.74	1.00	5.00	.93	2.45	0.76	1.00	4.50	.61
Leisure/recreation dysfunction	2.21	0.87	1.00	5.00	.79	2.29	0.87	1.00	5.00	.33
Satisfaction	2.46	0.54	1.06	4.42	---	2.47	0.61	1.00	5.00	.59
WHODAS-2 scales (number of items)	M	SD	Min.	Max.	Alpha	M	SD	Min.	Max.	Retest r
Cognition (6)	2.04	0.76	1.00	4.00	.86	1.95	0.78	1.00	5.00	.60
Mobility (5)	2.04	1.02	1.00	5.00	.91	2.07	1.02	1.00	5.00	.74
Self-care (4)	1.51	0.73	1.00	4.00	.83	1.49	0.77	1.00	5.00	.57
Getting along (5)	2.06	0.83	1.00	4.60	.76	2.03	0.91	1.00	5.00	.60
Life activities (8)	2.21	1.02	1.00	5.00	.94	2.21	1.05	1.00	5.00	.64
Participation (8)	2.23	0.88	1.00	5.00	.88	2.13	0.90	1.00	5.00	.67

Note. Session 1 N = 601. Session 2 n = 501 (LIFE-RIFT) or 497 (WHODAS-2). LIFE-RIFT = Longitudinal Interval Follow-up Evaluation Range of Impaired Functional Tool. WHODAS-2 = World Health Organization Disability Assessment Schedule 2.0. Alpha (internal consistency) and ICC (inter-rater) are reliability estimates. LIFE-RIFT satisfaction is a global rating made by participants, not interviewers.

The WHODAS-2 (World Health Organization, 2010) contains 36 self-report items rated using a 5-point Likert-type scale (none to extreme/cannot do) considering the past month. Items assess psychosocial impairment in understanding and communicating (cognition), getting around (mobility), self-care, getting along with people, life activities (e.g., household and work tasks), and participation in society (e.g., community activity involvement and enjoyment). Consistent with past research (Ustün et al., 2010), we found high internal consistency reliability for the WHODAS-2 scales (see Table 2).

Statistical Analyses

We quantified relations among measures using Pearson bivariate correlations and linear multiple regression. In all analyses, we reversed scores on IDAS-II well-being, the only scale on which higher scores indicate less psychopathology. We focused our conclusions on the magnitude and patterns findings, interpreting small, moderate, and large effect sizes for bivariate (r ≥ .10, .30, .50) and multiple (R² ≥ .02, .13, .26) correlations (Cohen, 1992). Secondarily, we considered statistical significance at p < .01, two-tailed.

We used Fisher’s transformation when computing the average correlation between symptom measures. We employed ordinary nonparametric bootstrapping to estimate the confidence interval (CI) for this average correlation, as well as for differences in average R² values between symptom and functioning measures. In particular, we used the percentile method to compute 99% CIs from 5000 random samples drawn with replacement.

About 17% of the data were missing at session 2, primarily due to attrition from the study. Considering data in Tables 1 and 2, participants with (n = 104) versus without (n = 497) missing session 2 data were quite similar. The two groups showed no session 1 differences in mean MINI, LIFE-RIFT, or WHODAS-2 scores, ps > .05. Similarly, only 2 of 19 IDAS-II scales differed between missing data groups at p < .05, and neither difference reached p < .01: Participants with session 2 missing data had slightly higher (effect size r) social anxiety (.10) and appetite loss (.10) at session 1. Because differences between groups were infrequent and small, we did not exclude participants with missing data from session 1 analyses or impute missing session 2 values.

Results

The sample’s symptomatology and psychosocial functioning scores generally indicated mild pathology, on average, with clear individual differences (see descriptive statistics in Tables 1–2). More specifically, averages on the IDAS-II scales fell at approximately the 55–82^nd (median = 72^nd) percentiles in a representative sample of adults in the United States (Nelson et al., 2017). Sample means in the top normative quartile included general depression, dysphoria, insomnia, appetite loss, ill temper, panic, claustrophobia, and traumatic intrusions. Proportions of patients reaching diagnostic thresholds on the MINI were: Major depressive disorder (29.8%), dysthymic disorder (19.0%), mania (17.8% lifetime), panic disorder (23.5%), agoraphobia (8.8%), social anxiety disorder (18.0%), obsessive-compulsive disorder (9.8%), post-traumatic stress disorder (8.8%), and generalized anxiety disorder (27.5%). Regarding psychosocial functioning at session 1, the current sample’s LIFE-RIFT total score (M = 9.59, SD = 2.28) was “fair” (between “good” and “poor” ranges; Judd et al., 2005) and more impaired than adults without psychopathology (M = 7.4; Judd et al., 2008). Finally, WHODAS 2.0 scale means were closest to the “mild impairment” rating of 2, with a total score (M = 12.27, SD = 4.12) at roughly the 75^th percentile of adult norms (World Health Organization, 2010).

Table 3 displays correlations of the IDAS-II scales with MINI symptom counts. The mean correlation was in the small-moderate range (.27, 99% CI [.24, .30]), and 17 of 19 IDAS-II scales showed at least one moderate or stronger (≥ .30) correlation, supporting our first hypothesis. All IDAS-II scales correlated statistically significantly with three or more MINI symptom counts.

Table 3.

Concurrent Correlations between the IDAS-II and MINI at Assessment Session 1

	|--------------------------------------------------MINI diagnostic symptom counts--------------------------------------------------|
IDAS-II scales	Major depressive disorder	Dysthymic disorder	Manic episode	Panic disorder	Agoraphobia	Social anxiety disorder	Obsessive-compulsive disorder	Post-traumatic stress disorder	Generalized anxiety disorder
General depression	.73	.61	.24	.38	.18	.29	.27	.36	.59
Dysphoria	.70	.59	.24	.36	.16	.30	.31	.35	.60
Lassitude	.50	.45	.18	.25	.08	.16	.13	.24	.42
Insomnia	.51	.41	.20	.30	.13	.15	.17	.23	.39
Suicidality	.46	.35	.16	.31	.11	.26	.19	.35	.33
Appetite loss	.36	.26	.15	.21	.11	.09	.17	.20	.23
Appetite gain	.19	.19	.08	.03	−.01	.07	.12	.06	.18
Ill-temper	.40	.36	.28	.21	.17	.17	.25	.22	.38
(low) Well-being	.46	.38	.07	.22	.10	.23	.06	.14	.32
Social anxiety	.43	.40	.24	.29	.20	.48	.23	.21	.41
Panic	.53	.46	.24	.41	.20	.19	.24	.33	.43
Claustrophobia	.34	.30	.24	.27	.28	.19	.31	.22	.27
Cleaning	.24	.22	.19	.18	.14	.17	.40	.16	.22
Checking	.26	.25	.22	.18	.19	.20	.45	.20	.25
Ordering	.13	.13	.15	.17	.07	.12	.41	.10	.15
Mania	.38	.34	.23	.30	.16	.19	.35	.27	.37
Euphoria	.04	.03	.22	.09	.05	−.05	.26	.15	.08
Traumatic intrusions	.47	.46	.26	.35	.19	.24	.26	.47	.46
Traumatic avoidance	.37	.33	.21	.30	.20	.24	.35	.36	.38

Note. N = 601. IDAS-II = Inventory of Depressive and Anxiety Symptoms-II. MINI = Mini International Neuropsychiatric Interview. Correlations ≥ |.11| are significant at p < .01, two-tailed. Correlations ≥ |.30| appear in bold type. The strongest correlation for each IDAS-II scale appears in italics. The strongest correlation for each MINI diagnostic symptom count is underlined.

Patterns of relations between the IDAS-II and MINI were readily understandable, in most instances. The IDAS-II general depression (.73), dysphoria (.70), lassitude (.50), insomnia (.51), suicidality (.46), ill-temper (.40), and (low) well-being (.46) scales showed their strongest correlations with MINI major depressive disorder, for example. However, IDAS-II panic (.53), claustrophobia (.34), and mania (.38) also correlated moderately to strongly with MINI major depressive disorder.

The IDAS-II panic scale showed a moderate correlation (.41) with panic disorder assessed with the MINI. In contrast, IDAS-II mania correlated statistically significantly but less strongly (.23) with MINI manic episode symptoms.

The IDAS-II social anxiety scale correlated most strongly with MINI social anxiety disorder (.48), whereas IDAS-II cleaning (.40), checking (.45), and ordering (.41) showed their strongest relations with MINI obsessive-compulsive disorder. Similarly, IDAS-II traumatic intrusions correlated most strongly with MINI post-traumatic stress disorder (.47), whereas IDAS-II traumatic avoidance showed relations with MINI generalized anxiety (.38) and major depressive disorder (.37) comparable to its correlation with post-traumatic stress disorder (.36).

The remaining two IDAS-II scales, appetite loss and euphoria, did not show moderate or stronger correlations (≥ .30) with any MINI diagnostic symptom count. Further, no IDAS-II scale showed a moderate or stronger correlation with MINI agoraphobia.

We next considered relations of the symptom measures (IDAS-II and MINI) with the functioning measures (LIFE-RIFT and WHODAS-2). As summarized in Table 4, the set of IDAS-II scales predicted psychosocial functioning moderately to strongly, both concurrently (session 1; mean R² = .24, range .13–.39) and prospectively (session 1 symptoms, session 2 functioning; mean R² = .22, range .11–.35). These omnibus relations supported our second and third hypotheses. Parallel values for the MINI predicting functioning concurrently (mean R² = .19, range .07–.36) and prospectively (mean R² = .15, range .06–.27) were somewhat weaker. All omnibus R² values were statically significant.

Table 4.

Omnibus Relations of Symptom with Functioning Measures

	Predictors: Session 1 symptoms
Outcomes:	IDAS-II scales		MINI symptom counts
Session 1 functioning	R2	Adjusted R2	R2	Adjusted R2
LIFE-RIFT
Work	.17	.15	.15	.14
Relationships	.15	.12	.09	.08
Leisure	.13	.10	.07	.06
Satisfaction	.24	.22	.20	.19
WHODAS-2
Cognition	.37	.35	.27	.26
Getting around	.21	.18	.17	.16
Self-care	.18	.15	.12	.11
Getting along	.34	.32	.27	.25
Life activities	.18	.16	.15	.13
Participation	.39	.37	.36	.35
Mean	.24	.21	.19	.17
Session 2 functioning
LIFE-RIFT
Work	.14	.11	.08	.06
Relationships	.11	.08	.07	.05
Leisure	.11	.07	.06	.04
Satisfaction	.22	.19	.14	.13
WHODAS-2
Cognition	.35	.32	.26	.24
Getting around	.20	.17	.18	.17
Self-care	.19	.16	.08	.06
Getting along	.30	.27	.21	.20
Life activities	.23	.20	.18	.17
Participation	.35	.32	.27	.26
Mean	.22	.19	.15	.14

Note. Session 1 N = 601. Session 2 n = 501 (LIFE-RIFT) or 497 (WHODAS-2). R² values obtained by regressing each functioning index on the set of 19 IDAS-II scales, or separately, on the set of 9 MINI symptom count variables. IDAS-II = Inventory of Depressive and Anxiety Symptoms-II. MINI = Mini International Neuropsychiatric Interview. LIFE-RIFT = Longitudinal Interval Follow-up Evaluation Range of Impaired Functional Tool. WHODAS-2 = World Health Organization Disability Assessment Schedule 2.0. All R² values p < .001. The strongest R² and Adjusted R² within ± .01 at each timepoint is underlined.

As shown in Table 5, the average incremental R² of the IDAS-II over the MINI was small-moderate for concurrent (mean ΔR² = .08, range .06–.12) and prospective (mean ΔR² = .09, range .07–.13) prediction of functioning. In contrast, the average increment of the MINI over the IDAS-II was quite small, both concurrently (mean ΔR² = .03, range .02–.05) and prospectively (mean ΔR² = .03, range .01–.07). Most (18/20) of the increments of the IDAS-II over the MINI were statistically significant, compared to less than half (8/20) of the increments of the MINI over the IDAS-II.

Table 5.

Incremental Prediction of Symptom from Functioning Measures

	IDAS-II increment to MINI			MINI increment to IDAS-II
Session 1 functioning	Δ R2	p	Δ Adj. R2	Δ R2	p	Δ Adj. R2
LIFE-RIFT
Work	.06	< .01	.04	.04	< .01	.03
Relationships	.07	< .01	.05	.02	.19	.01
Leisure	.08	< .01	.06	.03	.01	.02
Satisfaction	.08	< .01	.06	.04	< .01	.03
WHODAS-2
Cognition	.12	< .01	.10	.02	.01	.01
Getting around	.09	< .01	.06	.05	< .01	.04
Self-care	.08	< .01	.06	.03	.04	.01
Getting along	.11	< .01	.09	.03	< .01	.03
Life activities	.07	< .01	.05	.03	< .01	.02
Participation	.08	< .01	.06	.04	< .01	.04
Mean	.08		.06	.03		.02
Session 2 functioning
LIFE-RIFT
Work	.07	< .01	.04	.01	.75	−.01
Relationships	.07	.01	.03	.02	.39	.00
Leisure	.07	.01	.03	.02	.23	.01
Satisfaction	.11	< .01	.08	.03	.04	.01
WHODAS-2
Cognition	.13	< .01	.11	.04	< .01	.03
Getting around	.08	< .01	.05	.07	< .01	.05
Self-care	.13	< .01	.10	.02	.19	.01
Getting along	.11	< .01	.08	.02	.07	.01
Life activities	.08	< .01	.05	.03	.01	.02
Participation	.11	< .01	.08	.03	.01	.02
Mean	.09		.07	.03		.01

Note. Session 1 N = 601. Session 2 n = 501 (LIFE-RIFT) or 497 (WHODAS-2). Δ R² and Adjusted (Adj.) R² values were obtained by regressing on each functioning index the set of 9 MINI symptom count variables followed by the set of 19 IDAS-II scales (IDAS-II increment to MINI), or vice versa (MINI increment to IDAS-II). IDAS-II = Inventory of Depressive and Anxiety Symptoms-II. MINI = Mini International Neuropsychiatric Interview. LIFE-RIFT = Longitudinal Interval Follow-up Evaluation Range of Impaired Functional Tool. WHODAS-2 = World Health Organization Disability Assessment Schedule 2.0.

Because there were more IDAS-II scales (19) than MINI symptom counts (9), we also considered adjusted-R² values when comparing the measures. All omnibus (Table 4) and incremental (Table 5) adjusted-R² values were at least slightly higher for the IDAS-II than for the MINI, with one tie after rounding. The average omnibus adjusted R² with concurrent functioning for the IDAS-II (mean adjusted R² = .21) was moderate in size and greater than the MINI’s (mean adjusted R² = .17), difference = .04, 99% CI [.01, .09]. Similarly, the IDAS-II (mean adjusted R² = .19) was a stronger average prospective predictor of functioning than was the MINI (mean adjusted R² = .14), difference = .05, 99% CI [.02, .12]. In parallel, the IDAS-II’s average incremental adjusted R² was higher than the MINI’s increment when predicting functioning concurrently (IDAS-II mean Δ adjusted R² = .06, MINI mean Δ adjusted R² = .02) and prospectively (IDAS-II mean Δ adjusted R² = .07, MINI mean Δ adjusted R² = .01), although differences were small.¹

Tables 6 and 7 display concurrent (session 1) and prospective (session 1 symptoms, session 2 functioning) bivariate correlations, respectively, of the symptom measures with functioning indices. Most (16/19) IDAS-II scales, and just over half (5/9) of the MINI variables, showed as least one moderate or stronger concurrent correlation (r ≥ .30) with a functioning dimension (see Table 6). Symptom variables that often correlated substantively with poorer functioning included IDAS-II general depression (8 of 10 correlations ≥ .30), MINI major depressive disorder (8/10), IDAS-II dysphoria (7/10), and MINI dysthymic disorder (7/10). These patterns replicated prospectively, although moderate or stronger correlations were less frequent (Table 7).

Table 6.

Concurrent Correlations of Symptom with Functioning Measures at Assessment Session 1

	|----------------LIFE-RIFT----------------|				|----------------------------------WHODAS-2-----------------------------------|
IDAS-II scales	Work	Relationships	Leisure	Satisfaction	Cognition	Mobility	Self-care	Getting along	Life’s activities	Participation
General depression	.33	.27	.19	.41	.54	.32	.33	.48	.35	.58
Dysphoria	.31	.25	.19	.38	.55	.29	.32	.50	.34	.57
Lassitude	.36	.16	.11	.32	.38	.24	.25	.32	.36	.42
Insomnia	.15	.18	.07	.22	.31	.21	.23	.29	.14	.40
Suicidality	.18	.26	.11	.27	.35	.23	.32	.37	.18	.43
Appetite loss	.18	.17	.06	.18	.25	.20	.25	.25	.18	.33
Appetite gain	.06	.01	−.02	.05	.16	.10	.10	.12	.15	.14
Ill-temper	.16	.22	.12	.25	.39	.15	.26	.35	.22	.34
(low) Well-being	.19	.26	.28	.39	.31	.18	.14	.31	.19	.32
Social anxiety	.14	.21	.11	.22	.50	.22	.27	.52	.17	.47
Panic	.25	.22	.12	.25	.41	.34	.33	.35	.24	.48
Claustrophobia	.11	.17	.16	.16	.38	.34	.30	.33	.16	.39
Cleaning	.05	.18	.08	.12	.27	.21	.23	.28	.08	.30
Checking	.07	.10	.03	.09	.31	.15	.22	.29	.12	.28
Ordering	−.03	.09	−.03	−.01	.15	.09	.16	.20	.02	.17
Mania	.14	.17	.10	.19	.39	.16	.21	.30	.17	.36
Euphoria	.02	.08	.01	.02	.13	.08	.12	.13	.00	.12
Traumatic intrusions	.19	.23	.09	.25	.39	.23	.29	.38	.21	.43
Traumatic avoidance	.16	.24	.15	.24	.37	.25	.28	.35	.16	.40
MINI symptom counts
Major depressive disorder	.36	.28	.26	.43	.49	.38	.33	.43	.35	.55
Dysthymic disorder	.30	.21	.16	.34	.41	.34	.26	.33	.30	.48
Manic episode	.11	.14	.03	.12	.23	.13	.09	.20	.03	.23
Panic disorder	.12	.14	.05	.18	.21	.16	.15	.18	.12	.28
Agoraphobia	.04	.05	.05	.06	.17	.17	.13	.19	.13	.20
Social anxiety disorder	.05	.08	.01	.13	.22	.06	.04	.34	.04	.23
Obsessive-compulsive dis.	.05	.12	.07	.11	.21	.09	.13	.22	.08	.23
Post-traumatic stress dis.	.21	.19	.04	.25	.23	.19	.17	.24	.16	.32
Generalized anxiety dis.	.25	.21	.14	.32	.36	.25	.23	.37	.23	.45

Note. N = 601. IDAS-II = Inventory of Depressive and Anxiety Symptoms-II. MINI = Mini International Neuropsychiatric Interview. LIFE-RIFT = Longitudinal Interval Follow-up Evaluation Range of Impaired Functional Tool. WHODAS-2 = World Health Organization Disability Assessment Schedule 2.0. Correlations ≥ |.11| are significant at p < .01, two-tailed. Correlations ≥ |.30| appear in bold type. The strongest correlation within ± .01 of each LIFE-RIFT and WHODAS-2 index with the IDAS-II and MINI symptom counts, respectively, are underlined.

Table 7.

Predictive Correlations of Symptom (session 1) with Functioning (session 2) Measures

	|----------------LIFE-RIFT----------------|				|----------------------------------WHODAS-2-----------------------------------|
IDAS-II scales	Work	Relationships	Leisure	Satisfaction	Cognition	Mobility	Self-care	Getting along	Life’s activities	Participation
General depression	.29	.21	.20	.37	.51	.31	.24	.46	.41	.54
Dysphoria	.26	.19	.18	.34	.51	.28	.23	.48	.39	.52
Lassitude	.28	.13	.12	.28	.37	.22	.15	.31	.37	.40
Insomnia	.16	.12	.16	.18	.32	.24	.19	.29	.23	.42
Suicidality	.20	.17	.07	.22	.41	.23	.30	.37	.31	.36
Appetite loss	.15	.10	.05	.11	.25	.23	.22	.23	.21	.32
Appetite gain	.03	−.04	−.04	.02	.19	.09	.06	.15	.11	.13
Ill-temper	.12	.15	.09	.18	.37	.16	.20	.34	.25	.28
(low) Well-being	.15	.21	.23	.37	.23	.14	.02	.23	.19	.23
Social anxiety	.14	.18	.14	.20	.46	.20	.20	.48	.24	.41
Panic	.20	.17	.12	.23	.45	.37	.32	.36	.34	.47
Claustrophobia	.12	.15	.16	.13	.37	.29	.30	.32	.21	.35
Cleaning	.08	.14	.06	.05	.31	.22	.31	.28	.15	.24
Checking	.05	.05	.07	.07	.33	.16	.24	.29	.16	.31
Ordering	−.04	−.03	−.08	−.08	.20	.07	.21	.18	.04	.15
Mania	.18	.10	.09	.16	.41	.22	.27	.35	.26	.38
Euphoria	.05	.06	−.05	−.01	.23	.13	.25	.18	.09	.14
Traumatic intrusions	.24	.20	.13	.23	.37	.21	.23	.35	.25	.39
Traumatic avoidance	.14	.17	.11	.20	.31	.21	.25	.30	.16	.36
MINI symptom counts
Major depressive disorder	.24	.19	.19	.31	.46	.40	.21	.40	.41	.48
Dysthymic disorder	.23	.15	.22	.32	.42	.34	.24	.35	.33	.42
Manic episode	.05	.11	.01	.05	.21	.10	.06	.18	.11	.22
Panic disorder	.13	.15	.10	.18	.21	.19	.16	.17	.17	.25
Agoraphobia	.11	.06	.08	.05	.13	.19	.09	.11	.11	.15
Social anxiety disorder	.06	.12	.09	.15	.17	.07	.03	.24	.11	.18
Obsessive-compulsive dis.	.06	.13	.14	.12	.19	.10	.17	.22	.04	.19
Post-traumatic stress dis.	.19	.20	.11	.18	.28	.18	.16	.26	.19	.29
Generalized anxiety dis.	.17	.17	.18	.33	.38	.24	.16	.35	.29	.38

Note. N = 497–501. IDAS-II = Inventory of Depressive and Anxiety Symptoms-II. MINI = Mini International Neuropsychiatric Interview. LIFE-RIFT = Longitudinal Interval Follow-up Evaluation Range of Impaired Functional Tool. WHODAS-2 = World Health Organization Disability Assessment Schedule 2.0. Correlations ≥ |.12| are significant at p < .01, two-tailed. Correlations ≥ |.30| appear in bold type. The strongest correlation within ± .01 of each LIFE-RIFT and WHODAS-2 index with the IDAS-II and MINI symptom counts, respectively, are underlined.

Among the functioning variables, WHODAS-2 self-reported problems with cognition (e.g., attention, memory, communication), life participation (e.g., decreased activity and increased suffering due to poor health), and getting along (e.g., forming and maintaining social relationships) were the most frequent moderate or stronger correlates of the IDAS-II and MINI symptom measures, both concurrently (17/28, 19/28, and 16/28 correlations ≥ .30, respectively; Table 6) and prospectively (17/28, 16/28, and 14/28 correlations ≥ .30, respectively; Table 7). In contrast, interviewer-rated impairment in social relationships and leisure activities on the LIFE-RIFT showed many smaller, statistically significant, but not moderate or stronger (i.e., ≥ .30) correlations with IDAS-II or MINI symptom variables.

Discussion

In a sample of community-dwelling adults, results supported our hypotheses that the IDAS-II would converge with the MINI and predict psychosocial functioning both concurrently and prospectively. Overall, the IDAS-II was a stronger predictor of functioning than was the MINI. These findings bolster previous research showing clear convergence between the self-report IDAS-II with clinical interviews (Watson & O’Hara, 2017). More importantly, our findings expand validity evidence of the IDAS-II concerning relations with psychosocial functioning (De la Rosa-Cáceres et al., 2023; Santos et al., 2021). Based on the current findings, researchers and clinicians might consider using the IDAS-II to complement or replace potentially more time-consuming or expensive interview-based assessment of mood and anxiety disorder symptomatology, particularly when identifying potential sources of psychosocial impairment is a primary goal. Strengths and limitations of this approach are considered below.

Our findings contribute to the rapidly developing literature on the advantages of assessing psychopathology using a multidimensional system rather than with traditional polythetic diagnoses (Forbes et al., 2024; Kotov et al., 2022). The IDAS-II is a primary example of modern measures that reorganize heterogeneous DSM symptomatology into coherent dimensions, with hierarchical structures (Forbes et al., 2024; Kotov et al., 2022; Watson et al., 2022). In contrast to polythetic diagnoses and symptom counts, the IDAS-II assesses internally consistent symptom dimensions with well-established discriminant validity (Watson & O’Hara, 2017). Other multidimensional psychopathology measures’ incremental validity over traditional categorical diagnoses has been found for external variables including treatment utilization, treatment outcomes, and all-cause mortality (Kotov et al., 2022). The current analyses advance this literature by showing that IDAS-II scales outperformed MINI diagnostic symptom counts in predicting several dimensions of psychosocial functioning, both concurrently and prospectively.

The IDAS-II scale most consistently and substantively predicting psychosocial impairment was general depression. The IDAS-II general depression scale draws items from more specific symptom scales (e.g., IDAS-II dysphoria, insomnia, and [low] well-being), and in past research (Watson et al., 2007) has converged strongly with other broad-spectrum internalizing psychopathology measures, including the Beck Depression Inventory (Beck et al., 1996), Hamilton Rating Scale for Depression (Hamilton, 1960), and Beck Anxiety Inventory (Beck & Steer, 1990). A parallel pattern of results was observed in the current sample, in which the IDAS-II general depression scale correlated strongly with several diagnostic criterion counts, including major depressive, dysthymic, and generalized anxiety disorders, which in turn also related substantively to several psychosocial impairment dimensions. Thus, based on the current findings, researchers and clinicians seeking a traditional (i.e., broad but relatively non-specific) internalizing psychopathology measure would be well-served by IDAS-II general depression scale.

However, strengths of the IDAS-II also include assessment of more specific symptom dimensions that may clarify diagnoses and, moreover, help explain patterns of psychosocial functioning versus impairment. In the current study, all of the IDAS-II scales demonstrated statistically significant, and the majority of scales moderate to large, correlations with psychosocial functioning variables, both concurrently and prospectively. For example, difficulty with self-care measured by the WHODAS-2 related most clearly to suicidality and panic symptoms assessed with the IDAS-II. In contrast, dissatisfaction with one’s psychosocial functioning assessed with the LIFE-RIFT related substantively to IDAS-II dysphoria and low well-being.

To use the IDAS-II in clinical and research contexts, one viable strategy would be to give the full IDAS-II at baseline to inform diagnosis and treatment planning. Then, if it is not feasible to administer the entire IDAS-II repeatedly, use the subset of scales most relevant to specific research or treatment goals to assess changes over time. For example, the dysphoria scale could be used to capture general internalizing psychopathology, and one or two specific symptom scales (e.g., social anxiety, panic, or traumatic avoidance) with the greatest elevations at baseline could be selected to monitor treatment progress. Alternatively, the general depression scale could be used alone to index overall pathology, with specific item ratings monitored as needed (e.g., suicidality, insomnia; Stasik-O’Brien et al., 2019).

To interpret IDAS-II symptom scores normatively, data are available from a representative sample of United States adults (Nelson et al., 2018). In addition, cutoffs for describing symptom severity levels, and for screening (emphasizing sensitivity) or diagnostic (emphasizing specificity) purposes, have been proposed (Stasik-O’Brien et al., 2019). For example, IDAS-II dysphoria sum scores of 39 and 45 (i.e., average item ratings of 3.9 and 4.5 on the 1 = not at all to 5 = extremely scale), respectively, may indicate moderate (T = 65) and severe (T = 75) symptoms. In addition, a dysphoria score ≥ 22.5 may be useful in screening for major depressive disorder (sensitivity = .90, specificity = .63), whereas a dysphoria score ≥ 32.5 may help with diagnosis of major depressive disorder (sensitivity = .61, specificity = .91; Stasik-O’Brien et al., 2019).

Potential weaknesses of the IDAS-II scales observed here included no substantive relations to manic episode or agoraphobia symptoms assessed with the MINI. On the other hand, MINI agoraphobia and mania did not predict any psychosocial impairment index substantively, in the current sample of community-dwelling adults. In contrast, IDAS-II mania and IDAS-II claustrophobia (the strongest correlate of MINI agoraphobia) both predicted several areas of psychosocial impairment substantively, both concurrently and prospectively. We speculate that MINI manic episode or agoraphobia symptoms’ correlations with the IDAS-II and functioning measures may have been limited, in part, by the relative infrequency of more severe symptoms in the current sample. In addition, the difference in time frames for IDAS-II (past 2 weeks) versus MINI (lifetime history) mania may have limited the measures’ convergence and the MINI’s relations with recent functioning.

Two IDAS-II scales, euphoria and appetite gain, also showed no substantive relations with psychosocial functioning variables in the current sample. We again speculate that these scales’ correlations were limited by relatively mild current bipolar disorder symptomatology and appetite gain (e.g., euphoria, mania, and appetite gain sample means were within 0.3 SD of normative averages; Nelson et al., 2017). For example, in the symptomatically mild range, behaviors measured by the IDAS-II euphoria scale (e.g., high energy and self-esteem) may be relatively inert or double-edged, such that net effects on psychosocial functioning are small (e.g., forcefulness may have both costs and benefits in the workplace; cf. De Vries et al., 2020). Similarly, perhaps only pronounced appetite gain that leads to large body weight changes or obesity impairs functioning (cf. Vittengl, 2018). Future research is needed to test these speculations.

We observed post hoc that some dimensions of psychosocial functioning tended to correlate more strongly to anxiety and mood symptom measures. Two patterns emerged, although future research would be needed to test their replicability and our speculations about them. First, self-report versus interview ratings of impairment were often more strongly related to the symptom measures. Determining whether this pattern is limited to the specific self-report (WHODAS-2) and interview (LIFE-RIFT) functioning measures used in the current analyses, or is more general, requires additional research. If the pattern is more general, possible mechanisms include (a) individuals have richer knowledge of their functional impairment than can be conveyed fully during interviews, and/or (b) self-reported impairment is more highly loaded with non-specific emotional distress. The latter possibility better fit the second pattern we observed: Within instruments, functioning dimensions with more obvious emotional content tended to correlate more strongly with the symptom measures. For example, the LIFE-RIFT satisfaction-with-functioning index and the WHODAS-2 participation-in-society scale (e.g., “How much have you been emotionally affected…”) often related robustly with symptoms.

These general patterns provided context for interpretation of specific symptom measures’ correlations with functioning dimensions observed in the current dataset. For example, the IDAS-II social anxiety scale correlated slightly more highly with satisfaction with functioning (which has clear emotional content) than with relationship functioning (which has more obvious social content) on the LIFE-RIFT. However, these correlations were notably weaker than social anxiety’s correlations with the self-reported WHODAS-2 getting along (social content), cognition (also clear social content, such as starting and maintaining conversations) and participation (emotional content) scales.

Constraints on Generality

As in nearly all assessment and psychopathology research, limitations of the current study included the sample and measures. For example, results obtained from current sample of community-dwelling adults may not generalize fully to other populations, such as adolescents or mental health inpatients with more severe symptoms and functional impairment. In addition, we assessed psychosocial functioning with well-established interview and self-report measures. In what ways and to what extent our findings might have varied across other high-quality functioning measures is unknown. Finally, as a self-report measure, the IDAS-II may be limited by respondents’ lack of insight or bias in some behavioral ratings due to current emotion. Even so, the IDAS-II performed somewhat better than the MINI diagnostic interview in predicting psychosocial functioning, overall. Future research comparing the self-report IDAS-II with its clinician-rated counterpart (IDAS-CR; Watson et al., 2008; Watson et al., 2012) might improve understanding of the relative strengths and weaknesses of the two assessment modalities’ relations to psychosocial functioning.

Conclusions

Broad, specific, and practical measurement of mood and anxiety symptoms may inform diagnosis, treatment, outcome assessment and, moreover, help explain patterns of psychosocial impairment. Based on our results in a sample of community-dwelling adults, the self-report IDAS-II offers clinicians and researchers a valid means to pursue such measurement. We found that the IDAS-II converged with semi-structured interview assessment of mood and anxiety disorders and, as newly tested in this study, predicted a range of self-reported and interview-assessed psychosocial functioning dimensions both concurrently and prospectively.