Assessing distress in the community: psychometric properties and crosswalk comparison of eight measures of psychological distress

BACKGROUND

Many measures are available for measuring psychological distress in the community. Limited research has compared these scales to identify the best performing tools. A common metric for distress measures would enable researchers and clinicians to equate scores across different measures. The current study evaluated eight psychological distress scales and developed crosswalks (tables/figures presenting multiple scales on a common metric) to enable scores on these scales to be equated.

METHODS

An Australian online adult sample (N = 3620, 80% female) was administered eight psychological distress measures: Patient Health Questionnaire-4, Kessler-10/Kessler-6, Distress Questionnaire-5 (DQ5), Mental Health Inventory-5, Hopkins Symptom Checklist-25 (HSCL-25), Self-Report Questionnaire-20 (SRQ-20) and Distress Thermometer. The performance of each measure in identifying DSM-5 criteria for a range of mental disorders was tested. Scale fit to a unidimensional latent construct was assessed using Confirmatory Factor Analysis (CFA). Finally, crosswalks were developed using Item Response Theory.

RESULTS

The DQ5 had optimal performance in identifying individuals meeting DSM-5 criteria, with adequate fit to a unidimensional construct. The HSCL-25 and SRQ-20 also had adequate fit but poorer specificity and/or sensitivity than the DQ5 in identifying caseness. The unidimensional CFA of the combined item bank for the eight scales showed acceptable fit, enabling the creation of crosswalk tables.

CONCLUSIONS

The DQ5 had optimal performance in identifying risk of mental health problems. The crosswalk tables developed in this study will enable rapid conversion between distress measures, providing more efficient means of data aggregation and a resource to facilitate interpretation of scores from multiple distress scales.

Combined universal and selective prevention for adolescent alcohol use: a cluster randomized controlled trial

BACKGROUND

No existing models of alcohol prevention concurrently adopt universal and selective approaches. This study aims to evaluate the first combined universal and selective approach to alcohol prevention.

METHOD

A total of 26 Australian schools with 2190 students (mean age: 13.3 years) were randomized to receive: universal prevention (Climate Schools); selective prevention (Preventure); combined prevention (Climate Schools and Preventure; CAP); or health education as usual (control). Primary outcomes were alcohol use, binge drinking and alcohol-related harms at 6, 12 and 24 months.

RESULTS

Climate, Preventure and CAP students demonstrated significantly lower growth in their likelihood to drink and binge drink, relative to controls over 24 months. Preventure students displayed significantly lower growth in their likelihood to experience alcohol harms, relative to controls. While adolescents in both the CAP and Climate groups demonstrated slower growth in drinking compared with adolescents in the control group over the 2-year study period, CAP adolescents demonstrated faster growth in drinking compared with Climate adolescents.

CONCLUSIONS

Findings support universal, selective and combined approaches to alcohol prevention. Particularly novel are the findings of no advantage of the combined approach over universal or selective prevention alone.

The structure of adolescent psychopathology: a symptom-level analysis

BACKGROUND

Most empirical studies into the covariance structure of psychopathology have been confined to adults. This work is not developmentally informed as the meaning, age-of-onset, persistence and expression of disorders differ across the lifespan. This study investigates the underlying structure of adolescent psychopathology and associations between the psychopathological dimensions and sex and personality risk profiles for substance misuse and mental health problems.

METHOD

This study analyzed data from 2175 adolescents aged 13.3 years. Five dimensional models were tested using confirmatory factor analysis and the external validity was examined using a multiple-indicators multiple-causes model.

RESULTS

A modified bifactor model, with three correlated specific factors (internalizing, externalizing, thought disorder) and one general psychopathology factor, provided the best fit to the data. Females reported higher mean levels of internalizing, and males reported higher mean levels of externalizing. No significant sex differences emerged in liability to thought disorder or general psychopathology. Liability to internalizing, externalizing, thought disorder and general psychopathology was characterized by a number of differences in personality profiles.

CONCLUSIONS

This study is the first to identify a bifactor model including a specific thought disorder factor. The findings highlight the utility of transdiagnostic treatment approaches and the importance of restructuring psychopathology in an empirically based manner.

Psychometric evaluation of the DSM-IV criterion B mania symptoms in an Australian national sample

BACKGROUND

Although numerous studies have examined the latent structure of major depression, less attention has focused on mania. This paper presents the first investigation outside the USA to evaluate the psychometric properties of the DSM-IV criterion B mania symptoms using item response theory (IRT).

METHOD

Data were drawn from the Australian 2007 National Survey of Mental Health and Well-Being (NSMHWB, n = 8841). The psychometric performance of the mania symptoms was evaluated using a two-parameter logistic model. Because substance use disorders (SUDs) frequently co-occur with mania and can influence manic symptom expression, differential item functioning (DIF) between mania respondents with/without a SUD diagnosis was also assessed.

RESULTS

Factor analysis supported a unidimensional trait underlying mania. The grandiosity symptom displayed the highest discrimination whereas discrimination was lowest for decreased need for sleep. Relatively speaking, grandiosity tapped the severe end and increased goal-oriented activities tapped the mild end of the mania severity continuum. The symptoms generally performed equivalently between those with/without a SUD diagnosis, with one exception; the activities with painful consequences symptom was endorsed at lower levels of severity, and hence more frequently, by those with a SUD diagnosis versus those without a SUD diagnosis.

CONCLUSIONS

Accurate conceptualization of latent structure has crucial theoretical, statistical and clinical implications. The symptoms generally performed well in distinguishing between respondents with differing levels of liability, but others did not, suggesting modification is warranted to ensure optimal use in epidemiological samples. Given the dearth of psychometric evaluation studies of mania, further research replicating these results is necessary.