Date of this Version
Background In 2017, the QIC-WD began working with the Louisiana Department of Children and Family Services (DCFS). At that time DCFS had an average turnover rate of 24% across the state with one region having a rate of 51% and another with a low of 8%. A thorough needs assessment identified that high caseloads and a large number of administrative tasks were barriers to caseworkers’ ability to effectively engage in the clinical aspects of their work and led to the decision to redesign the frontline caseworker job. The intervention created a new job, the Child Welfare Team Specialist (CWTS), responsible for administrative tasks and reorganized workers into Permanency (former Foster Care) and Prevention teams. Implementation began in July 2019 and rolled out in two additional groups in November 2019 and February 2020. The evaluation included an implementation evaluation and a quasi-experimental pretest-posttest nonequivalent groups design comparing workers providing investigative, in-home, and foster care services in three parishes implementing the job redesign (the experimental group) to a group of workers providing the same services in six similar parishes doing business as usual (the comparison group). The goals of the evaluation were to determine whether job redesign reduced overload and increased worker well-being, increased time spent on clinical work with children and families resulting in improved practice, and decreased turnover as outlined in the site logic model and theory of change. Workforce Demographics Data were collected in May 2020 from 270 caseworkers in the study sample, 78 from the redesign group and 192 from comparison parishes. Of these survey respondents 94% were female,72% were African American, 22% were White, 2% were Hispanic, 4% identified as multi-racial or other, and less than 1% identified as Asian or as Indigenous or Pacific Islander. The average age was 38 years. The majority (58%) had bachelor’s degrees. Another 41% had master’s degrees. Approximately 34% had social work degrees. There were no significant differences between experimental and comparison groups related to gender, race, or age. Respondents in the comparison group reported a higher percentage of master’s degrees than those in the experimental group. Of 18 CWTS, 83% were female, 78% were African American and 22% were White; 78%, and all who responded had bachelor’s degrees in fields other than social work. The average age was 35 years. Evaluation Findings Focus group feedback indicated that participants were satisfied with the model and wanted to see it maintained in some form, particularly the role of the CWTS. Participants cited a wide range of benefits for the workforce including less stress and improved morale. They pointed to improvements in timeliness and quality of the work and favorable impacts on children and families, including more timely referrals, earlier service provision, faster case closure, and reductions in cases going to family services or foster care. Surveys were administered in May 2020, three months after full implementation of the redesign, and again in November 2021. Caseworkers in the experimental group scored significantly higher than caseworkers in the comparison group on measures of team cohesion, worklife balance, fit with the organization, and fit with their work group, and significantly lower than the Frontline Job Redesign Key Findings Louisiana Department of Children and Family Services 2 | July 2023 key findings comparison group on role conflict. Expected group differences in self-efficacy, fit with the demands of the job, organizational commitment, work engagement, job involvement, job embeddedness, burnout and role overload did not materialize and there were no significant differences in job satisfaction or intent to stay or leave the agency at either time point. While there were no significant group differences in perceptions of supervisor support or competence on either survey, experimental group participants rated their supervisors more highly on planning and organizing work -unit activities on both the initial and follow-up surveys. Early results of the redesign found that caseworkers in the experimental group rated their work stress significantly lower than the comparison group, however, group differences in ratings of work stress were no longer significant on the follow-up survey in November 2021, which could be related to COVID-19 restrictions and major hurricanes affecting both the experimental and comparison parishes. Following the implementation of the redesign (May 2020), caseworkers in the experimental group rated their jobs significantly lower on job complexity, degree of specialization required, and variety of skills needed than they had in an earlier assessment conducted as part of the needs assessment (in 2018). This suggests that those in redesigned teams perceived their jobs to be more streamlined, however, these differences were not maintained in November 2021, which may be a result of virtual work arrangements. Time study data indicated increases in the percentage of time caseworkers in redesigned units spent on clinical activities, such as interviews, home visits, and case planning, and that the new CWTSs were taking on a wide range of administrative tasks, such as opening and closing cases in the data systems, obtaining consents and archiving records. Specifically, between July 2019 and October 2021, workers in Prevention Units providing investigative and short-term in-home services increased the proportion of time spent on clinical aspects of the job from 46% to 65% and time spent on administrative aspects of the job decreased from 53% to 33%. In the Permanency Units, time spent on clinical aspects of the work increased from 55% to 61%, while time spent on administrative tasks decreased from 45% to 39%. Random samples of cases served by the experimental and comparison groups were evaluated using a customized tool to examine aspects of the work expected to be positively impacted by the job redesign, such as getting services to families more quickly, making transfers to in-home or foster care services more seamless, and improving the quality and timeliness of services to children in out-of-home care and their parents. Reviews showed statistically significant improvements for the experimental group from baseline to full implementation in accuracy of risk and safety assessments; early provision of prevention services; timely involvement of the family services worker in cases identified as requiring in-home family services; and quality and frequency of contacts with children and parents. Significant improvements occurred in diligent efforts to locate absent parents, and timely notifications to participants in Family Team Meetings which were part of the CWTS’s duties. Changes in early provision of prevention services; timely involvement of the family services workers and timeliness of notifications to families were significantly greater for the experimental group than for the comparison group. Between the rollout of the first redesigned units in July 2019 and the state’s move to virtual work in March 2020, admissions to foster care in the experimental parishes decreased, while admissions rose in the comparison parishes. The difference in foster care placements (compared to what would have been expected if the experimental parishes followed the same trend as the comparison parishes) was statistically significant and translated to 103 fewer families experiencing a foster care admission. In the eight months following the move to virtual work the gap Workers in Prevention and Permanency Teams increased time spent on clinical tasks as a result of the redesign. between the expected and actual levels of placements for the experimental group narrowed. DCFS reported that cases accepted for investigation during this time were more serious, making it more difficult to safely maintain children in their homes. Recurrence rates during this time period did not increase suggesting that the lower rates of foster care placements in the experimental parishes were not offset by a rise in subsequent maltreatment. Based on supervisor ratings of behaviors associated with risk of turnover, in May 2020 workers in the experimental group were significantly less likely to exhibit elevated risk than workers in the comparison group. Specifically, the experimental group was significantly less likely than the comparison group to leave early, expend less effort, be less productive, show less focus, show less interest in clients, contribute less in meetings, show less enthusiasm for the agency mission, do only the minimum amount of work, and show less interest in pleasing and more dissatisfaction with their supervisor. However, there were no significant group differences in supervisor’s assessments of their workers’ risk of turnover on the November 2021 survey. Analysis of group differences in actual turnover rates before and after implementation were inconclusive given the small number of data points available and the degree of variation in rates by state fiscal year. Although not all hypothesized changes in worker wellbeing, job satisfaction, and commitment were supported, the job redesign showed promise in reducing perceptions of work stress and role conflict, increasing time available for clinical work with families, and improving case practice. Additional research is needed with respect to intervention effects on turnover. The Team This project would not have been possible without the partnership of DCFS. They identified a statewide team that represented all of the regions and key positions in the organization. They committed to and participated in monthly meetings over multiple years to plan and support the initiative and its evaluation. They also provided essential data and valuable insight throughout the project. We greatly appreciate the work of all members of the steering committee, the training unit, Louisiana Civil Service, human resources, and all of the staff who worked to make this project a success. We are especially indebted to Marketa Garner Walters for having the vision and courage to commit the agency to undertaking a job redesign which had never been done in Child Welfare before and to Rhenda Hodnett and Leslie Calloway for their leadership and dedication to this project. The QIC-WD would also like to extend a special thanks to Brandy Malatesta, Shelly Johnson, Anthony Ellis, Connie Guillory, Fertaeshia Broussard, Leslie Breaux, Melissa Thompson, Leslie Lyons, and Stacey Mire for their many contributions, and to Elizabeth Reveal and Ryan Dodge for their work in providing and interpreting the administrative data used in the evaluation.