The psychological assessment of depression: unipolars versus bipolars

Identifying major predictors for parenting stress in a caregiver of autism spectrum disorder using machine learning models

Introduction

Previous studies have investigated predictive factors for parenting stress in caregivers of autism spectrum disorder (ASD) patients using traditional statistical approaches, but their study settings and results were inconsistent. Herein, this study aimed to identify major predictors for parenting stress in this population by developing explainable machine learning models.

Methods

Study participants were collected from the Department of Child and Adolescent Psychiatry, Severance Hospital, Yonsei University College of Medicine, Seoul, the Republic of Korea between March 2016 and October 2020. A total of 36 model features were used, which include subscales of the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) for caregivers' psychopathology, Social Responsiveness Scale-2 for core symptoms, and Child Behavior Checklist (CBCL) for behavioral problems. Machine learning classifiers [eXtreme Gradient Boosting (XGBoost), random forest (RF), logistic regression, and support vector machine (SVM) classifier] were generated to predict severe total parenting stress and its subscales (parental distress, parent-child dysfunctional interaction, and difficult child). Model performance was assessed by area under the receiver operating curve (AUC), sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. We utilized the SHapley Additive exPlanations tree explainer to investigate major predictors.

Results

A total of 496 participants were included [mean age of ASD patients 6.39 (SD 2.24); 413 men (83.3%)]. The best-performing models achieved an AUC of 0.831 (RF model; 95% CI 0.740-0.910) for parental distress, 0.814 (SVM model; 95% CI 0.720-0.896) for parent-child dysfunctional interaction, 0.813 (RF model; 95% CI 0.724-0.891) for difficult child, and 0.862 (RF model; 95% CI 0.783-0.930) for total parenting stress on the test set. For the total parenting stress, ASD patients' aggressive behavior and anxious/depressed, and caregivers' depression, social introversion, and psychasthenia were the top 5 leading predictors.

Conclusion

By using explainable machine learning models (XGBoost and RF), we investigated major predictors for each subscale of the parenting stress index in caregivers of ASD patients. Identified predictors for parenting stress in this population might help alert clinicians whether a caregiver is at a high risk of experiencing severe parenting stress and if so, providing timely interventions, which could eventually improve the treatment outcome for ASD patients.

A study of remitted and treatment-resistant depression using MMPI and including pessimism and optimism scales

Background

The psychological aspects of treatment-resistant and remitted depression are not well documented.

Methods

We administered the Minnesota Multiphasic Personality Inventory (MMPI) to patients with treatment-resistant depression (n = 34), remitted depression (n = 25), acute depression (n = 21), and healthy controls (n = 64). Pessimism and optimism were also evaluated by MMPI.

Results

ANOVA and post-hoc tests demonstrated that patients with treatment-resistant and acute depression showed similarly high scores for frequent scale (F), hypochondriasis, depression, conversion hysteria, psychopathic device, paranoia, psychasthenia and schizophrenia on the MMPI compared with normal controls. Patients with treatment-resistant depression, but not acute depression registered high on the scale for cannot say answer. Using Student's t-test, patients with remitted depression registered higher on depression and social introversion scales, compared with normal controls. For pessimism and optimism, patients with treatment-resistant depression demonstrated similar changes to acutely depressed patients. Remitted depression patients showed lower optimism than normal controls by Student's t-test, even though these patients were deemed recovered from depression using HAM-D.

Conclusions

The patients with remitted depression and treatment-resistant depression showed subtle alterations on the MMPI, which may explain the hidden psychological features in these cohorts.

The nature and treatment of depression in bipolar disorder: A review and implications for future psychological investigation

Bipolar depression is poorly understood and researched, yet it is has a huge impact on functioning in bipolar disorder. This review explores the current status of research regarding the phenomenology, natural history, neuropsychology, psychosocial predictors and cognitive style of bipolar depression. The current status of pharmacotherapy and psychological treatment of bipolar depression is also described. In particular, the manner in which cognitive behaviour therapy for bipolar depression has been adapted from CBT for unipolar depression is critically evaluated. It is concluded that there appears to be a considerable overlap between the features of unipolar and bipolar depression, yet there is also emerging evidence for specific elements. The ability of current psychological theories of bipolar disorder to account for the findings are compared, and as a consequence, a new preliminary integrative model is proposed to direct future hypothesis-led research, which will need to incorporate more suitable populations and utilise more objective methods of assessment.

Linking Personality Disorders and Clinical Syndromes on the MCMI-III

We examined the relationship between personality disorders (PDs) and clinical syndromes (CSs) as measured by the Millon Clinical Multiaxial Inventory-III (MCMI-III; Millon, 1997) in a large, heterogeneous sample of psychiatric patients (N = 2,366) who completed the instrument as part of routine assessment following presentation for treatment. Using separate sets of base rate (BR) and nonoverlapping scale scores, we factor analyzed the PD and CS scales together and then separately. We correlated results from the latter analyses to determine how trait dimensions were associated with syndrome dimensions. We also studied co-occurrence at the scale level by examining CS score profiles of patients who were grouped according to their highest PD scale elevation > or = BR75. Results for the two score sets were very similar and were consistent with previous research on the MCMI-III and its predecessors that identified 3 underlying dimensions loading both PD and CS scales. Three fourths (76.2%) of the sample had a highest PD scale > or = BR75, and among these, 90% had at least 1 CS scale > or = BR75, whereas 62.4% had 3 or more CS scales above this elevation. Findings underscore the substantial overlap between PDs and CSs along 3 dimensions that resemble Horney's (1945) tripartite interpersonal distinction of moving toward, away, and against, as well as Eysenck's (1994) higher order factors of neuroticism, extraversion, and psychoticism.

Depression in inpatients: bipolar vs unipolar

162 depressed inpatients were divided into three diagnostic groups to compare patterns of sociodemographic characteristics, psychopathology, and psychosocial: 35 had a single episode of major depression, 96 had recurrent major depression, and 31 had a bipolar disorder. Psychopathology and psychosocial functioning were measured by clinician-rated scales, Montgomery-Asberg Depression Rating Scale, Hamilton Rating Scale for Depression, Clinical Global Impression, and self-rating scales, Symptom Checklist-90, Social Support Questionnaire, Social Adjustment Scale. The three groups were comparable on sociodemographic variables, with the exception of education. Univariate analyses showed a similar social impairment as measured by Social Support Questionnaire, Social Adjustment Scale, and no significant differences were recorded for the psychopathology when the total test scores (Montgomery-Asberg Depression Rating Scale, Hamilton Rating Scale for Depression, Clinical Global Index, Symptom Checklist-90) were evaluated. Some differences emerged for single items in the Montgomery-Asberg Depression Rating Scale and Symptom Checklist-90. These findings suggest a substantial similarity among the three groups. Results are discussed in terms of the clinical similarities between unipolar and bipolar patients during a depressive episode as well as the limitations of cross-sectional study implies.

Personality and affect characteristics of outpatients with depression

This investigation was designed to examine the relationship between depression severity and personality disorders measured by the Millon Clinical Multiaxial Inventory-II (Millon, 1987) and affectivity measured by the Positive Affectivity/Negative Affectivity Schedule (Watson, Clark, & Tellegen, 1988). Discriminant analyses were employed to identify the personality and affective dimensions that maximally discriminate between 4 different levels of depressive severity. Differences between the 4 levels of depressive severity are suggestive of unique patterns of personality characteristics. Discriminant analysis showed that 74.8% of the cases were correctly classified by a single linear discriminant function, and that 61% of the variance in depression severity was accounted for by selected personality and affect variables. Results extend current conceptualizations of comorbidity and are discussed with respect to depression severity.

Assessing depression with the MMPI and MMPI-2

The usefulness of the MMPI (Hathaway & McKinley, 1951 ) and MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) for diagnosing and assessing symptomatic depression has been the subject of considerable debate for a number of years. In this article, we review the relative contributions of the MMPI and MMPI-2 clinical and content scales in predicting depression. Positive predictive power, negative predictive power, and overall classification rate were computed for Scale 2 (D) of the MMPI and MMPI-2 and the Depression content scale (DEP) of the MMPI-2. Scale 2 (D) of both the MMPI and MMPI-2 appears to be moderately accurate in predicting depression. Although some studies suggest that the content scale DEP provides incremental validity over Scale 2 (D) of the MMPI-2, the results of this review indicate that the content scale DEP of the MMPI-2 does not exceed the diagnostic efficiency of Scale 2 in predicting depression.

The use of the MMPI-2 for the assessment of depressive and psychotic disorders

Considering the normative changes incorporated into the MMPI-2 and the introduction of content scales, this study examined its usefulness for the diagnosis and assessment of depression and psychosis in a heterogeneous sample of 264 psychiatric inpatients. We examined the mean group profiles and diagnostic efficiency of single scales at specified cutoff scores for these conditions. We also conducted cross-validated stepwise regression using all the basic and content scales as well as hierarchical regression examining the incremental validity of the basic and content scales. In general, the MMPI-2 profiles were found to be sensitive to group differences and the derived regression equations proved to be stable and fairly good at classification; but single scales were less useful for diagnosis. Changes in norms made the MMPI-2 more specific than the MMPI, and the introduction of new content scales offered considerable additional clinical information and incremental validity.