Using Human Resources Data to Predict Turnover of Community Mental Health Employees: Prediction and Interpretation of Machine Learning Methods.

This study used machine learning (ML) to predict mental health employees' turnover in the following 12 months using human resources data in a community mental health centre. The data contain 621 employees' information (e.g., demographics, job information and client information served by employees) hired between 2011 and 2021 (56.5% turned over during the study period). Six ML methods (i.e., logistic regression, elastic net, random forest [RF], gradient boosting machine [GBM], neural network and support vector machine) were used to predict turnover, along with graphical and statistical tools to interpret predictive relationship patterns and potential interactions. The result suggests that RF and GBM led to better prediction according to specificity, sensitivity and area under the curve (>0.8). The turnover predictors (e.g., past work years, work hours, wage, age, exempt status, educational degree, marital status and employee type) were identified, including those that may be unique to the mental health employee population (e.g., training hours and the proportion of clients with schizophrenia diagnosis). It also revealed nonlinear and nonmonotonic predictive relationships (e.g., wage and employee age), as well as interaction effects, such that past work years interact with other variables in turnover prediction. The study indicates that ML methods showed the predictability of mental health employee turnover using human resources data. The identified predictors and the nonlinear and interactive relationships shed light on developing new predictive models for turnover that warrant further investigations.

Why do Stayers Stay? Perceptions of White and Black Long-Term Employees in a Community Mental Health Center.

Previous research has focused on factors influencing turnover of employees in the mental health workforce, yet little research has explored reasons why employees stay. To facilitate retaining a diverse mental health workforce, the current study aimed to elucidate factors that contributed to employees' tenure at a community mental health center (CHMC) as well as compare these perceptions between Black and White employees. Long-term employees (7 years or more) from one urban CMHC (n = 22) completed semi-structured stayer interviews. Using emergent thematic analysis, stayer interviews revealed four major themes for why they have stayed at the organization for 7 years or more: (1) work as a calling, (2) supportive relationships, (3) opportunities for growth or meaningful contribution, and (4) organization mission's alignment with personal attributes or values. Comparison between Black and White stayer narratives revealed differences in their perceptions with work as a calling and opportunities for growth and meaningful contribution. Guided by themes derived from stayer interviews, the current study discusses theoretical (e.g., job embeddedness theory, theory of racialized organizations, self-determination theory) and practical implications (e.g., supporting job autonomy, Black voices in leadership) in an effort to improve employee retention and address structural racism within a mental health organization.

Trauma-informed climate scale-10 and attitudes related to trauma-informed care-10: An examination of scale psychometrics using data from disability service providers.

BACKGROUND: Assessment is fundamental to the implementation of trauma-informed care. As trauma-informed care advances among organizations supporting people with intellectual and developmental disabilities (IDD), it is critical that they have access to validated assessment tools. OBJECTIVE: This is the first study to examine the psychometric properties of two brief measures associated with trauma-informed care, Trauma-Informed Climate Scale (TICS-10) and Attitudes Related to Trauma-Informed Care (ARTIC-10), using data from the IDD service industry. METHODS: We employed structural equation modeling to examine the factor structure, reliability of TICS-10 and ARTIC-10, and construct validity with the ProQOL using secondary data from 374 service providers. RESULTS: We confirmed the factorial validity of TICS-10 and ARTIC-10 with single factor solutions; however, modifications were necessary to achieve adequate model fit. CONCLUSIONS: The current study provides initial evidence of the validity and reliability of TICS-10 and ARTIC-10 when used within organizations supporting people with IDD. Recommendations for subsequent improvements and future research of the scales are provided.

Machine Learning with Human Resources Data: Predicting Turnover among Community Mental Health Center Employees.

BACKGROUND: Human resources (HR) departments collect extensive employee data that can be useful for predicting turnover. Yet, these data are not often used to address turnover due to the complex nature of recorded data forms. AIMS OF THE STUDY: The goal of the current study was to predict community mental health center employees' turnover by applying machine learning (ML) methods to HR data and to evaluate the feasibility of the ML approaches. METHODS: Historical HR data were obtained from two community mental health centers, and ML approaches with random forest and lasso regression as training models were applied. RESULTS: The results suggested a good level of predictive accuracy for turnover, particularly with the random forest model (e.g., Area Under the Curve was above .8) compared to the lasso regression model overall. The study also found that the ML methods could identify several important predictors (e.g., past work years, wage, work hours, age, job position, training hours, and marital status) for turnover using historical HR data. The HR data extraction processes for ML applications were also evaluated as feasible. DISCUSSION: The current study confirmed the feasibility of ML approaches for predicting individual employees' turnover probabilities by using HR data the organizations had already collected in their routine organizational management practice. The developed approaches can be used to identify employees who are at high risk for turnover. Because our primary purpose was to apply ML methods to estimate an individual employee's turnover probability given their available HR data (rather than determining generalizable predictors at the wider population level), our findings are limited or restricted to the specific organizations under the study. As ML applications are accumulated across organizations, it may be expected that some findings might be more generalizable across different organizations while others may be more organization-specific (idiographic). IMPLICATIONS FOR HEALTH CARE PROVISION AND USE: The organization-specific findings can be useful for the organization's HR and leadership to evaluate and address turnover in their specific organizational contexts. Preventing extensive turnover has been a significant priority for many mental health organizations to maintain the quality of services for clients. IMPLICATIONS FOR HEALTH POLICIES: The generalizable findings may contribute to broader policy and workforce development efforts. IMPLICATIONS FOR FURTHER RESEARCH: As our continuing research effort, it is important to study how the ML methods and outputs can be meaningfully utilized in routine management and leadership practice settings in mental health (including how to develop organization-tailored intervention strategies to support and retain employees) beyond identifying high turnover risk individuals. Such organization-based intervention strategies with ML applications can be accumulated and shared by organizations, which will facilitate the evidence-based learning communities to address turnover. This, in turn, may enhance the quality of care we can offer to clients. The continuing efforts will provide new insights and avenues to address data-driven, evidence-based turnover prediction and prevention strategies using HR data that are often under-utilized.

Introduction to the special section on racial disparities in psychiatric rehabilitation services.

Although racial disparities in psychiatric rehabilitation services are not new, the urgency of systematic approaches to address them has gained increased attention. In particular, the current social and political climate has spotlighted historically persistent and universally prevalent problems in equitable care. This special section, consisting of six studies and a Letter to the Editor, reveals the operation and impact of structural racism and highlights the need for race-conscious practice and research in psychiatric rehabilitation. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Cognitive and Affective Empathy as Indirect Paths Between Heterogeneous Depression Symptoms on Default Mode and Salience Network Connectivity in Adolescents.

Depression amongst adolescents is a prevalent disorder consisting of heterogeneous emotional and functional symptoms-often involving impairments in social domains such as empathy. Cognitive and affective components of empathy as well as their associated neural networks (default mode network for cognitive empathy and salience network for affective empathy) are affected by depression. Depression commonly onsets during adolescence, a critical period for brain development underlying empathy. However, the available research in this area conceptualizes depression as a homogenous construct, and thereby miss to represent the full spectrum of symptoms. The present study aims to extend previous literature by testing whether cognitive and affective empathy indirectly account for associations between brain network connectivity and heterogeneous depression symptoms in adolescents. Heterogeneous functional and emotional symptoms of depression were measured using the child depression inventory. Our results indicate that cognitive empathy mediates the association between default mode network functional connectivity and emotional symptoms of depression. More specifically, that adolescents with a stronger positive association between the default mode network and cognitive empathy show lower emotional depression symptoms. This finding highlights the importance of cognitive empathy in the relationship between brain function and depression symptoms, which may be an important consideration for existing models of depression in adolescents.

Structural Racism, Workforce Diversity, and Mental Health Disparities: A Critical Review.

OBJECTIVE: Racial workforce diversity has been suggested as a critical pathway to address persistent racial mental health disparities. However, structural racism has been noted to diminish such workforce diversity efforts. The purpose of this critical review is to identify the mechanisms through which structural racism operates in organizations, including mental health organizations, to undermine workforce diversity efforts and reinforce inequities. METHODS: Using the theories of racialized organizations, the current review critically draws on literature underscoring the racial character of organizations as mezzo-level racialized structures that may systematically activate and uphold white privilege in the mental health workplace. RESULTS: Findings suggest that in the context of institutionalized white dominance, workers of color within mental health organizations may experience race-based cultural exclusion, identity threat, and racialized workplace emotional expression, and be burdened by racialized tasks. The workers of color may also become the means for organizations to attract communities of color due to their diverse characteristics, yet workers' effects to address disparities in mental health are minimized due to potential racialized organizational forces, including the whiteness of organizational leadership and color-blindness. CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: Structural racism may create resistance to the efforts and effects of a racially diverse workforce within mental health organizations. This review calls for a race-conscious framework that drastically shifts the traditional organizational structure to an inverted hierarchy (i.e., client-centered management) to maximize diversity efforts in the mental health organizational workforce to address racial disparities in mental health.

Combining proration and full information maximum likelihood in handling missing data in Likert scale items: A hybrid approach.

Large-scale surveys are common in social and behavioral science research. Missing data often occur at item levels due to nonresponses or planned missing data designs. In practice, the item scores are typically aggregated into scale scores (i.e., sum or mean scores) for further analyses. Although several strategies to handle item-level missing data have been proposed, most of them are not easy to implement, especially for applied researchers. Using Monte Carlo simulations, we examined a practical hybrid approach to deal with item-level missing data in Likert scale items with a varying number of categories (i.e., four, five, and seven) and missing data mechanisms. Specifically, the examined approach first uses proration to calculate the scale scores for a participant if a certain proportion of item scores is available (a cutoff criterion of proration) and then use full information maximum likelihood to deal with missing data at the scale level when scale scores cannot be computed due to the selected proration cutoff criterion. Our simulation results showed that the hybrid approach was generally acceptable when the missing data were randomly spread over the items, even when they had different thresholds/means and loadings, with caution to be taken when the missingness is determined by one of the scale items. Based on the results, we recommend using the cutoff of 30% or 40% for proration when the sample size is small and the cutoff of 40% or 50% when the sample size is moderate or large.

Process research: compare and contrast the recovery-orientated strengths model of case management and usual community mental health care.

BACKGROUND: The strengths model of case management (SMCM), which was developed by Rapp and Goscha through collaborative efforts at the University of Kansas, assists individuals with mental illness in their recovery by mobilizing individual and environmental resources. Increasing evidence has shown that the utilization of the SMCM improves outcomes, including increased employment/educational attainment, reduced hospitalization rates, higher self-efficacy, and hope. However, little is known about the processes through which the SMCM improves outcomes for mental health service users. This study explores the views of case workers and service users on their experience of providing or receiving the SMCM intervention. METHODS: A qualitative design was employed using individual interviews with service users and case workers drawn from two study conditions: the SMCM group and the control group (treatment as usual). For both study conditions, service users were recruited by either centres-in-charge or case workers from integrated community centres for mental wellness (ICCMWs) operated by three non-governmental organizations (NGOs) in different districts of Hong Kong. Through purposeful sampling, 24 service users and 14 case workers from the SMCM and control groups joined the study. We used an inductive approach to analyse the qualitative data. RESULTS: We identified two overarching themes: service users' and case workers' (1) perceptions of the impacts of the interventions (SMCM and control group) and (2) experiences of the interventions, such as features of the interventions and the factors that facilitated the outcomes. The results showed that there were improvements in the functional recovery of the SMCM group in areas such as employment and family relationships, how self-identified goals were achieved, and how service users gained a better understanding of their own strengths and weaknesses. Regarding their experience of the interventions, participants in both the SMCM group and the control group reported that a good relationship between service users and case workers was vital. However, some concerns were raised about the use of SMCM tools, including the strengths assessment and the personal recovery plan (PRP) and the risk of case workers being subjective in the presentation of cases in group supervision sessions. CONCLUSION: The results were promising in terms of supporting the use of the SMCM, with some refinements, in mental health services for Chinese clients. TRIAL REGISTRATION: The Australian New Zealand Clinical Trials Registry (ANZCTR), ACTRN12617001435370 , registered on 10/10/2017.

Network functional connectivity underlying dissociable cognitive and affective components of empathy in adolescence.

Empathy, the capacity to understand and share others' emotions, can occur through cognitive and affective components. These components are different conceptually, behaviorally, and in the brain. Neuroimaging task-based research in adolescents and adults document that cognitive empathy associates with the default mode and frontoparietal networks, whereas regions of the salience network underlie affective empathy. However, cognitive empathy is slower to mature than affective empathy and the extant literature reveals considerable developmental differences between adolescent and adult brains within and between these three networks. We extend previous work by examining empathy's association with functional connectivity within and between these networks in adolescents. Participants (n = 84, aged 13-17; 46.4% female) underwent resting state fMRI and completed self-report measures (Interpersonal Reactivity Index) for empathy as part of a larger Nathan-Kline Institute study. Regression analyses revealed adolescents reporting higher cognitive empathy had higher within DMN connectivity. Post hoc analysis revealed cognitive empathy's association within DMN connectivity is independent of affective empathy or empathy in general; and this association is driven by positive pairwise connections between the bilateral angular gyri and medial prefrontal cortex. These results suggest introspective cognitive processes related to the DMN are specifically important for cognitive empathy in adolescence.

Assessing the Recovery Assessment Scale Across Time.

OBJECTIVE: The Recovery Assessment Scale (RAS) is one of the most used recovery measures in recovery-oriented practice evaluation of people with mental health conditions. Although its psychometric properties have been extensively studied, one critical piece of information that is missing from the literature is evidence of its longitudinal factorial invariance-that is, whether the RAS measures the same recovery construct across time. The authors empirically tested the longitudinal factorial invariance assumption for the RAS. METHODS: Structural equation modeling was used to test the longitudinal factorial invariance of the RAS with data longitudinally obtained at three time points from 167 people with severe mental illness. RESULTS: The longitudinal factorial invariance assumption was supported (i.e., configural, metric, partial scalar, factor variance and covariance invariance). CONCLUSIONS: This study found empirical evidence that the RAS can measure the same recovery construct over time and thus meets one of the important prerequisites for longitudinal assessment.

Factors that affect quality of care among mental health providers: Focusing on job stress and resources.

Objective: High-quality, person-centered care is a priority for mental health services. The current study conducted secondary data analysis to examine the impact of job stress (i.e., interaction with high-risk consumer cases, increased caseload, emotional exhaustion) and resources (i.e., increased organizational and supervisory support, autonomy, role clarity) on providers' perceived quality of care. Methods: Data consisted of 145 direct care providers from an urban community mental health center. Structural equation modeling was used for testing the hierarchical regression model, sequentially adding job stress and resource variables in the prediction models for the quality of care (i.e., person-centered care, discordant care [conflict with consumers and tardiness]). Results: Person-centered care was positively associated with increased role clarity, organizational support, and larger caseload size, while a lower level of discordant care was associated with lower emotional exhaustion, smaller caseload size, less interaction with high-risk consumer cases, and with increased role clarity. Conclusions and Implications for Practice: Resources on the job may be particularly important for improved person-centered care, and lowering job stress may help reduce discordant care. The current study suggests the need for the mental health organizations to attend to both job stress and resources for providers to improve the quality of care. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Longitudinal Effects of Cognitive and Affective Empathy on Adolescent Substance Use.

Background: A deficit in either socio-cognitive or socio-affective components of empathy is associated with the severity of substance use by late adolescence. What remains unknown is how longitudinal changes in these components of empathy predict adolescent substance using behavior. Methods: This secondary data analysis used data that followed adolescents in outpatient treatment for substance use (n = 826) during treatment and at 6 months post-treatment. To examine cross-lagged effects of empathy on substance use over time, we used a latent basis growth curve model. Results: Increases in affective empathy predicted reduced substance use over time. However, cognitive empathy did not predict substance use after controlling for other covariates. Conclusions: Lower levels of affective empathy may indicate a developmental vulnerability for substance using behavior. Modifying affective empathy may be a viable treatment target for reducing adolescent substance use.

Characteristics and Job Stressors Associated With Turnover and Turnover Intention Among Community Mental Health Providers.

OBJECTIVE: This study aimed to examine the provider characteristics and job stressors that are related to turnover intention and actual turnover among community mental health providers. METHODS: Secondary analyses were conducted with data collected from 186 community mental health providers from two agencies. Self-reported provider characteristics, job stressors, and turnover intention data were collected with the baseline survey, and actual turnover data were obtained from the agencies 12 months later. Bivariate analyses were conducted to examine factors associated with each turnover variable. RESULTS: Turnover intention and actual turnover were correlated, yet a distinct set of variables was associated with each outcome. Namely, job stressors were related to turnover intention, while provider characteristics were related to actual turnover. CONCLUSIONS: Given that both turnover intention and actual turnover have important implications for both providers and agencies, it is critical to consider differential factors associated with each.

Study protocol for a randomised controlled trial evaluating the effectiveness of strengths model case management (SMCM) with Chinese mental health service users in Hong Kong.

INTRODUCTION: Strengths-based approaches mobilise individual and environmental resources that can facilitate the recovery of people with mental illness. Strengths model case management (SMCM), developed by Rapp and Goscha through collaborative efforts at the University of Kansas, offers a structured and innovative intervention. As evidence of the effectiveness of strengths-based interventions come from Western studies, which lacked rigorous research design or failed to assure fidelity to the model, we aim to fill these gaps and conduct a randomised controlled trial (RCT) to test the effectiveness of SMCM for individuals with mental illness in Hong Kong. METHODS AND ANALYSIS: This will be an RCT of SMCM. Assuming a medium intervention effect (Cohen's d=0.60) with 30% missing data (including dropouts), 210 service users aged 18 years or above will be recruited from three community mental health centres. They will be randomly assigned to SMCM groups (intervention) or SMILE groups (control) in a 1:1 ratio. The SMCM groups will receive strengths model interventions from case workers, whereas the SMILE groups will receive generic care from case workers with an attention placebo. The case workers will all be embedded in the community centres and will be required to provide a session with service users in both groups at least once every fortnight. There will be two groups of case workers for the intervention and control groups, respectively. The effectiveness of the SMCM will be compared between the two groups of service users with outcomes at baseline, 6 and 12 months after recruitment. Functional outcomes will also be reported by case workers. Data on working alliances and goal attainment will be collected from individual case workers. Qualitative evaluation will be conducted to identify the therapeutic ingredients and conditions leading to positive outcomes. Trained outcome assessors will be blind to the group allocation. ETHICS AND DISSEMINATION: Ethical approval from the Human Research Ethics Committee at the University of Hong Kong has been obtained (HRECNCF: EA1703078). The results will be disseminated to service users and their families via the media, to healthcare professionals via professional training and meetings and to researchers via conferences and publications. TRIAL REGISTRATION NUMBER: 12617001435370; Pre-results.

Mediational paths from supervisor support to turnover intention and actual turnover among community mental health providers.

OBJECTIVE: Turnover is a critical problem for community mental health providers, and supervisors may play an important role in mitigating turnover. The current study examined the potential impact of supervisory support on turnover intention and actual turnover among community mental health providers. METHOD: We conducted path analyses with data collected longitudinally from 186 direct clinical care providers at two community mental health centers. RESULTS: Increased supervisory support was associated with lower turnover intention 6 months later, as well as reduced actual turnover 12 months later. Type of supervisory support mattered: supervisors' emotional support was most strongly associated with reduced turnover intention and turnover. However, client goal alignment support (supporting providers to help consumers achieve their goals) was directly associated with increased turnover. CONCLUSIONS AND IMPLICATIONS FOR PRACTICE: The current study suggests an important role for direct supervisors to attend to care providers' emotional support needs, which may reduce turnover intention and actual turnover. Differential supervisory support functions might impact turnover in unique ways; thus, examining more detailed change mechanisms would facilitate our understanding of factors that may prevent future turnover. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Impact of Supervisory Support on Turnover Intention: The Mediating Role of Burnout and Job Satisfaction in a Longitudinal Study.

High rates of provider turnover are problematic for our mental health system. Research indicates that supervisory support could alleviate some turnover intention by decreasing emotional exhaustion (a key component of burnout) as well as by increasing job satisfaction. However, the potential mediation mechanisms have not been rigorously tested. Longitudinal data collected from 195 direct clinical care providers at two community mental health centers identified positive effects of supervisory support on reduced turnover intention through reduced emotional exhaustion. Job satisfaction was not a significant mediator. Supervisory support may help mitigate turnover intention through work-related stress reduction.

Measuring Quality of Care in Community Mental Health: Validation of Concordant Clinician and Client Quality-of-Care Scales.

Measuring quality of care can transform care, but few tools exist to measure quality from the client's perspective. The aim of this study was to create concordant clinician and client self-report quality-of-care scales in a sample of community mental health clinicians (n = 189) and clients (n = 469). The client scale had three distinct factors (Person-Centered Care, Negative Staff Interactions, and Inattentive Care), while the clinician scale had two: Person-Centered Care and Discordant Care. Both versions demonstrated adequate internal consistency and validity with measures related to satisfaction and the therapeutic relationship. These measures are promising, brief quality assessment tools.

A Comparative Effectiveness Trial to Reduce Burnout and Improve Quality of Care.

Clinician burnout is presumed to negatively impact healthcare quality; yet scant research has rigorously addressed this hypothesis. Using a mixed-methods, randomized, comparative effectiveness design, we tested two competing approaches to improve care-one addressing clinician burnout and the other addressing how clinicians interact with consumers-with 192 clinicians and 469 consumers at two community mental health centers. Although qualitative reports were promising, we found no comparative effectiveness for either intervention on burnout, patient-centered processes, or other outcomes. Discussion includes identifying ways to strengthen approaches to clinician burnout.

Improvements in Irritability with Open-Label Methylphenidate Treatment in Youth with Comorbid Attention Deficit/Hyperactivity Disorder and Disruptive Mood Dysregulation Disorder.

OBJECTIVE: The purpose of this open-label study was to examine the effects of long-acting methylphenidate (MPH) treatment on irritability and related emotional symptoms associated with disruptive mood dysregulation disorder (DMDD) in youth with comorbid attention-deficit/hyperactivity disorder (ADHD). METHODS: The sample included 22 medication-free male and female subjects (ages 9-15) who met criteria for both DMDD and ADHD. Participants underwent a 4-week trial of long-acting MPH treatment (Concerta®), with weekly dosing increases until a therapeutic dose was reached. Repeated measures t-tests were used to compare pre- and posttreatment ratings of primary and secondary measures. The primary outcome was self-report irritability. Secondary outcomes included parent and child ratings of emotional frequency, emotional lability, and negative affect (NA). Multiple regression was used to examine the impact baseline hyperactivity, age, gender, race, socioeconomic status, or comorbid diagnosis had on treatment outcomes. RESULTS: Significant improvements (medium to large effect sizes) in child-rated irritability as well as parent and child ratings of emotional lability, NA, and anger were found. As anticipated, ADHD symptoms also improved. While a majority of the sample saw improvement in child-rated irritability (71%), symptoms worsened a small proportion (19%), and an even smaller portion experienced no change (10%). No demographics, psychiatric comorbidities, or severity of ADHD symptoms influenced treatment outcomes. CONCLUSIONS: Study findings suggest that MPH treatment significantly improved mood and emotional symptoms associated with DMDD comorbid with ADHD. These findings, coupled with good tolerability in this open-label pilot study supports further research into the use of MPH as a first-line treatment for DMDD. Future work examining MPH treatment of youth with DMDD with and without comorbid ADHD is needed.