Psychometric comparison of single-item, short, and comprehensive depression screening measures in Korean young adults

Criterion validity of a single-item measure of fear avoidance behavior following mild traumatic brain injury

Maladaptive coping such as fear avoidance behavior can prolong recovery from mild traumatic brain injury (mTBI). Routine assessment of fear avoidance may improve management of mTBI. This study aimed to validate a single-item measure of fear avoidance to make its assessment more pragmatic. The present study is a secondary analysis of a clinical trial that involved adults with persistent post-concussion symptoms (N = 90, 63% female). Participants completed the single-item fear avoidance rating, a validated legacy measure of fear avoidance (Fear Avoidance Behavior after Traumatic Brain Injury; FAB-TBI), and measures of anxiety (General Anxiety Disorder-7), depression (Patient Health Questionnaire-9), post-concussion symptoms (Rivermead Postconcussion Symptoms Questionnaire), and disability (World Health Organization Disability Assessment Schedule 12.0). Questionnaires were completed twice, at baseline (mean 18.1 weeks post injury) and again 12-16 weeks later following study-delivered rehabilitation in addition to usual care. We analyzed the associations (Spearman's correlations) and agreement (weighted Kappa) between the single-item and FAB-TBI at baseline, posttreatment, individual FAB-TBI item scores, and the change in scores between baseline and posttreatment. In addition, we examined correlations between the single-item fear avoidance measure and related constructs, including anxiety, depression, post-concussion symptoms, and disability. The single-item fear avoidance measure correlated strongly with the FAB-TBI both at baseline and following treatment (ρ = 0.63 - 0.67, p < .001), and moderately with FAB-TBI item scores (ρ = 0.4 - 0.6). The correlation between the change in these scores from baseline to posttreatment was moderate (ρ = 0.45, p < .001). Agreement between the single-item fear avoidance measure and discretized FAB-TBI scores was moderate (κ = 0.45 - 0.51). Before and after treatment, the single-item fear avoidance measure correlated moderately with anxiety (ρ = 0.34), depression (ρ = 0.43), post-concussion symptoms (ρ = 0.50), and disability (ρ = 0.43). The FAB-TBI was more strongly correlated with these measures (ρ = 0.53 - 0.73). In summary, the present study supports the criterion validity of the single-item fear avoidance measure. This measure may be a useful screening and monitoring tool for patients with mTBI but is not a substitute for the FAB-TBI questionnaire.

Effect of a smartphone-based physical intervention on depression, fitness factors and movement characteristics in adults

BACKGROUND

Physical activity has been shown to correlate with mental health and a reduction in symptoms of depression. However, the majority of research has focused only on the effects of either aerobic or nonaerobic exercise on depressive symptoms, while the use of novel technological innovations such as mobile phone-based activity programs and their effects on movement characteristics are underrepresented. This study had two objectives: (1) to investigate how effectively 4 weeks of mobile phone-based physical activity can affect depressive scores (CES-10-D and PHQ) and fitness levels and (2) to investigate the whether 4 weeks of mobile phone-based physical activity affected participants' movement characteristics.

METHODS

A total of 31 participants were included and divided into an exercise group (n = 21) and a control group (n = 10). The exercise group was instructed to use a mobile phone-based exercise program 5 times per week for 4 weeks. Pre- and post-exercise, the participants' depression score (CES-10-D, PHQ9), fitness level (YMCA, grip strength) and movement characteristics (postural sway, movement ROM, movement speeds, etc.) for three Azure Kinect physical activity games based on different fitness factors (balance game, cardiovascular game, reaction game) were measured.

RESULTS

Mixed model ANOVA revealed significant differences between pre- and post-intervention depression scores on the PHQ9 (P = .001) and CES-10-D (P < .001) in both the exercise group and the control group, but not between groups. In terms of movement characteristics, there was an increase in body sway (P = .045) and vertical head movement (P = .02) in the cardiovascular game jogging condition for the exercise group. In the reaction game, the exercise group showed a significant reduction in the number of mistakes (P = .03). There were no other significant differences for the other variables.

CONCLUSION

The results revealed no differences in the reduction in depression scores between the exercise group and the control group. However, this study showed that a mobile phone-based physical activity intervention affects in-game movement characteristics such as body sway and vertical head movement and therefore may show the potential of using activity-promoting mobile games for improving movement.

The role of cytokines and Indolamine-2.3 dioxygenase in experiencing a psycho-neurological symptom cluster in hematological cancer patients: IL-1alpha, IL-1beta, IL-4, IL-6, TNF-alpha, kynurenine, and tryptophan

OBJECTIVE

This study examined (a) whether there are a subgroup of cancer patients experiencing the selected psycho-neurological symptoms as a cluster (depression, cognitive impairment, fatigue, sleep disturbance, and pain); (b) whether demographic and clinical characteristics and pro-inflammatory cytokines (IL-1α, IL-1β, IL-4, IL-6, TNF-alpha) are associated with subgroup membership; and (c) whether the activity of indolamine-2.3 dioxygenase(IDO) is associated with pro-inflammatory cytokine activity and psycho-neurological symptom cluster experience.

METHODS

This was a prospective cohort study where 149 hematologic patients were recruited from a university hospital and 65 healthy volunteers provided control data. Latent profile analyses were conducted to identify subgroups at two time points: the last day of chemotherapy and 1 week after chemotherapy completion. Influencing factors of subgroup membership were examined by logistic regression.

RESULTS

A substantial number of patients (33%, 34% at each time point) experienced the selected psycho-neurological symptoms as a cluster. Older age and elevated IL-1α and IL-6 were associated with experiencing the psycho-neurological symptom cluster. IDO activity was higher in the patients experiencing psycho-neurological symptom cluster; and was positively associated with IL-6. Symptom severity, IL-1α, IL-6, and IDO activity were all significantly higher in cancer patients than in the healthy controls. The findings were preserved across time points.

CONCLUSIONS

The activation of pro-inflammatory cytokines and their cross-talk with IDO may be a common biological mechanism, underlying a psycho-neurological symptom cluster experience. The novel approaches for symptom assessment and management can be developed by assessing multiple psycho-neurological symptoms as a cluster and by targeting their common biological pathway.

Designing and Using Surveys in Nursing Research: A Contemporary Discussion

This commentary summarizes the contemporary design and use of surveys or questionnaires in nursing science, particularly in light of recent reporting guidelines to standardize and improve the quality of survey studies in healthcare research. The benefits, risks, and limitations of these types of data collection tools are also briefly discussed.

Roles of biological and psychosocial factors in experiencing a psychoneurological symptom cluster in cancer patients

PURPOSE

(a) To identify subgroups with unique psychoneurological symptom-cluster experience (depression, cognitive impairment, fatigue, sleep disturbance, and pain) and (b) to examine whether the selected demographic, clinical, psychological, and biological factors determine a symptom-cluster experience in cancer patients.

METHOD

The sample included 203 patients with diverse cancer types recruited from a Korean university hospital. Latent profile analyses were conducted to identify subgroups. Influencing factors of subgroup membership (demographic/clinical variables, hemoglobin level, social support, and psychological stress) were included as covariates in latent profile analysis and analyzed by multinomial logistic regression.

RESULTS

Latent profile analyses classified patients into two subgroups with a unique symptom cluster experience: patients experiencing high intensity in all symptoms within the cluster (the all-high-symptom subgroup, 71%) and patients experiencing low intensity in all symptoms within the cluster (all-low-symptom subgroup, 29%). The validity of the two subgroups was confirmed by the group classification accuracy (97% of the all-low-symptom subgroup and 99% of the all-high-symptom subgroup) and by significant Wald's mean equality tests, showing each symptom (depression, cognitive impairment, fatigue, sleep disturbance, and pain) significantly differentiated the two subgroups (ps < .001). Psychological stress independently determined the subgroup membership. Patients with high levels of stress were more likely to be in the all-high-symptom group (OR = 4.69, p < .0001). Hemoglobin level, cancer diagnosis, social support, and previous chemotherapy experience did not influence group membership.

CONCLUSIONS

A large number of patients experience five psychoneurological symptoms simultaneously due to psychological stress. Interventions targeted to stress would be beneficial for those patients.

Smallest detectable change and test-retest reliability of a self-reported outcome measure: Results of the Center for Epidemiologic Studies Depression Scale, General Self-Efficacy Scale, and 12-item General Health Questionnaire

RATIONALE, AIMS, AND OBJECTIVES

This study aims to examine the smallest detectable change (SDC) and test-retest reliability of the Center for Epidemiologic Studies Depression Scale (CES-D), General Self-Efficacy Scale (GSES), and 12-item General Health Questionnaire (GHQ-12).

METHOD

We tested 154 young adults at baseline and 2 weeks later. We calculated the intra-class correlation coefficients (ICCs) for test-retest reliability with a two-way random effects model for agreement. We then calculated the standard error of measurement (SEM) for agreement using the ICC formula. The SEM for agreement was used to calculate SDC values at the individual level (SDC_ind ) and group level (SDC_group ).

RESULTS

The study participants included 137 young adults. The ICCs for all self-reported outcome measurement scales exceeded 0.70. The SEM of CES-D was 3.64, leading to an SDC_ind of 10.10 points and SDC_group of 0.86 points. The SEM of GSES was 1.56, leading to an SDC_ind of 4.33 points and SDC_group of 0.37 points. The SEM of GHQ-12 with bimodal scoring was 1.47, leading to an SDC_ind of 4.06 points and SDC_group of 0.35 points. The SEM of GHQ-12 with Likert scoring was 2.44, leading to an SDC_ind of 6.76 points and SDC_group of 0.58 points.

CONCLUSION

To confirm that the change was not a result of measurement error, a score of self-reported outcome measurement scales would need to change by an amount greater than these SDC values. This has important implications for clinicians and epidemiologists when assessing outcomes.

Psychosocial interventions to improve mental health in adults with vision impairment: systematic review and meta-analysis

PURPOSE

To systematically assess the literature on psychosocial interventions to improve mental health (i.e. depression, anxiety, mental fatigue, loneliness, psychological stress and psychological well-being) in visually impaired adults (≥18 years).

METHODS

The databases Medline, Embase and Psychinfo were searched for relevant studies, which were categorised into randomised controlled trials (RCTs), non-RCTs and before and after comparisons (BA). The Cochrane Collaboration Risk of Bias Tool was used to assess study quality. Standardised mean differences (SMD) were calculated to quantitatively summarise the outcomes of the RCTs and non-RCTs in a meta-analysis. Meta-regression was used to explore sources of heterogeneity in the data.

RESULTS

The search identified 27 papers (published between 1981 and 2015), describing the outcomes of 22 different studies (14 RCTs, four non-RCTs, and four BAs). Pooled analyses showed that interventions significantly reduced depressive symptoms (SMD -0.30, 95% confidence interval (CI) -0.60 to -0.01), while effects on anxiety symptoms, mental fatigue, psychological stress and psychological well-being were non-significant. Meta-regression analyses showed homogeneity in effect sizes across a range of intervention, population, and study characteristics. Only a higher age of participants was associated with less effective results on depressive symptoms (b = 0.03, 95% CI 0.01 to 0.05), psychological stress (b = 0.07, 95% CI 0.01 to 0.13) and psychological well-being (b = -0.03, 95% CI -0.05 to 0.01). However, after removing a clear outlier the overall effect on depressive symptoms and the influence of age on depressive symptoms and psychological stress were no longer significant, while the influence of age on psychological well-being remained.

CONCLUSIONS

There is currently only limited evidence for the effectiveness of psychosocial interventions in the field of low vision. More well-designed trials are needed with specific attention for interventions tailored to the needs of elderly patients.

Measurement of fatigue: Comparison of the reliability and validity of single-item and short measures to a comprehensive measure

BACKGROUND

Evidence is needed on the clinicometric properties of single-item or short measures as alternatives to comprehensive measures.

OBJECTIVES

We examined whether two single-item fatigue measures (i.e., Likert scale, numeric rating scale) or a short fatigue measure were comparable to a comprehensive measure in reliability (i.e., internal consistency and test-retest reliability) and validity (i.e., convergent, concurrent, and predictive validity) in Korean young adults.

METHODS

For this quantitative study, we selected the Functional Assessment of Chronic Illness Therapy-Fatigue for the comprehensive measure and the Profile of Mood States-Brief, Fatigue subscale for the short measure; and constructed two single-item measures. A total of 368 students from four nursing colleges in South Korea participated. We used Cronbach's alpha and item-total correlation for internal consistency reliability and intraclass correlation coefficient for test-retest reliability. We assessed Pearson's correlation with a comprehensive measure for convergent validity, with perceived stress level and sleep quality for concurrent validity and the receiver operating characteristic curve for predictive validity.

RESULTS

The short measure was comparable to the comprehensive measure in internal consistency reliability (Cronbach's alpha=0.81 vs. 0.88); test-retest reliability (intraclass correlation coefficient=0.66 vs. 0.61); convergent validity (r with comprehensive measure=0.79); concurrent validity (r with perceived stress=0.55, r with sleep quality=0.39) and predictive validity (area under curve=0.88). Single-item measures were not comparable to the comprehensive measure.

CONCLUSIONS

A short fatigue measure exhibited similar levels of reliability and validity to the comprehensive measure in Korean young adults.

Reliability and Validity of the 4-Item Version of the Korean Perceived Stress Scale

This study was designed to evaluate the reliability (internal consistency, reproducibility over time, and measurement precision) and validity (factorial, convergent, and concurrent validity) of the 4-item Korean Perceived Stress Scale in comparison to the 10-item Korean Perceived Stress Scale. Participants were 650 students from seven schools of nursing in South Korea. Measures included Korean versions of the Perceived Stress Scale, the Pittsburgh Sleep Quality Index, and a single-item quality-of-life measure. The 4-item Perceived Stress Scale had lower Cronbach alpha than the 10-item Perceived Stress Scale (.65 vs. .80), but its item-total correlations (.35-.47) and split-half coefficient (.81) were comparable to the 10-item Perceived Stress Scale (item-total correlations = .20-.56; split-half coefficient = .82). Reproducibility of the 4-item scale after a 1-week interval (r = .67, ICC absolute agreement = .64) was similar to the 10-item scale (r = .70, ICC absolute agreement = .65). The standard error of measurement was slightly higher in the 4-item scale. Scale constructs of the two were identical in exploratory factor analysis. The two scales were highly correlated (r = .93) and showed similar levels of concurrent validity with sleep quality and quality of life. In conclusion, the 4- and 10-item Korean Perceived Stress Scale are similar in psychometric qualities. © 2016 Wiley Periodicals, Inc.