Texas Decreased Caseworker Turnover. The QIC-WD is Implementing and Testing Some of Their Strategies and Building Evidence

The QIC-WD is dedicated to building evidence. Foundational work in the early years of the Center included combing the literature from multiple fields (e.g., business, education, industrial-organizational psychology) to inform strategies to strengthen the child welfare workforce. In 2018, Casey Family Programs reported on how the Texas Department of Family and Protective Services (DFPS) reduced caseworker turnover by implementing multiple changes to system operations (How did Texas decrease caseworker turnover and stabilize its workforce?). This post discusses how the QIC-WD is working with eight other jurisdictions and building on what was learned in Texas.

According to the report, Texas initiated multiple changes at once, including:

• new DFPS leadership;
• building champions in the legislature and governor’s office;
• pay increases of $12,000 per front-line worker;
• an increase in the size of the workforce (by 8% in 2017 and 5% in 2018 – 2019);
• changes to the training and onboarding processes, such as, balancing classroom training with immediate field experience, and graduated caseloads;
• new hiring protocols including an aptitude screening tool and realistic job preview;
• mentorship; and
• attention to agency culture by addressing human resources (HR) practices, implementing recognition ceremonies, standardizing the supervisor selection process, and addressing the performance management system.

There is some evidence to suggest that many of these strategies can enhance employee recruitment and retention, however, it is difficult to disentangle the effect of any of them when they occur simultaneously. Texas’ approach to tackling multiple challenges at once lowered turnover over the course of one year (a 27.5% decrease from September 2016 – September 2017), and reports indicate turnover continued to decline in 2018 (by 5% - 7% across various positions).  While these strategies, in the aggregate, had a positive impact on turnover what happens if jurisdictions implement one or just some of these changes?

The table below lists the strategies implemented in Texas and by the QIC-WD sites. The QIC-WD is a time-limited, research project. Therefore, we do not have the funding to test pay raises or the capacity to increase the size of the child welfare workforce in our sites. (The Texas legislature invested $292 million in 2017 to hire 828 more employees and another $88 million in 2018 – 2019 to hire 597 employees.) However, we are studying some of the same types of strategies used in Texas to answer questions about their effectiveness as stand-alone interventions.

One lesson documented in the Texas report was that “leadership support is necessary.” The QIC-WD agrees and is building on this implementation support. QIC-WD sites must have agency leadership (from child welfare and human resources, as appropriate) as part of the site implementation team and most sites identified leadership and managers as champions of the selected interventions. Just as Texas is undergoing leadership changes (a new director was appointed in October 2019) we have also seen changes in agency and local project leadership in six of our eight sites (75%) since the project began in 2017. Our implementation study  tracks leadership changes and our sites have reported on the importance of buy-in from the top as well as staff at all levels of the organization when initiating workforce changes. Furthermore, there may be other questions the QIC-WD can help answer such as the level of involvement needed and how to sustain change when leadership changes.

In addition to the importance of leadership to successful implementation, other elements of implementation science are part of the QIC-WD model. For example, QIC-WD sites created cross-sectional teams that meet frequently to assess data, find the best solution to problems in the organization, and to shepherd the change from installation to full implementation. Referred to as implementation teams, they also include experts from the QIC-WD to study both the process and outcomes of the workforce strategy, feed the evaluation findings back to the system for continuous quality improvement, and keep enthusiasm high.  We believe that when implementation teams are used appropriately, changes are installed and practiced with higher degrees of fidelity. This can lead to effective changes being sustained over time.  We are eager to follow the news and findings related the changes in Texas and we will monitor the QIC-WD sites to see how long changes last.

The Casey report recognizes that a variety of the strategies Texas implemented may aid organizational culture change. Organizational culture is defined as the way things are done in an organization. In the Texas example, they made staffing changes, centralized their personnel system, implemented early coaching and counseling for staff, formalized staff recognition, made changes in the supervisor selection process, and did some right-sizing of their child welfare caseloads. All these elements shape organizational culture. Only one QIC-WD site is specifically addressing organizational culture and climate through the implementation of an intervention: the Division of Milwaukee Child Protective Services. Other sites, however, are addressing some of the aspects of organizational culture. In Oklahoma, for example, they are redesigning the selection process, in Ohio, they are training and coaching supervisors, in Louisiana they are reorienting the work around teaming and a focus on prevention, and in Virginia and Washington they are using technology to change when, where and how the work gets done in child welfare.

The success of Texas’ ability to stabilize their workforce should be celebrated and learned from. What we are doing as the QIC-WD, is testing complex, but somewhat discrete, workforce interventions that were chosen as a result of a thorough needs assessment process that matched the intervention with drivers of turnover in that jurisdiction. This way we have interventions carefully grounded in the needs of the site, and we can test the impact of the intervention on outcomes like turnover, performance, and child outcomes.  We are also using implementation science to guide our work so that the interventions are deeply embedded in the organizations where the testing is taking place. This will increase the chance that the interventions will work. We are also tracking the impact of interventions and pieces of the interventions over time with longitudinal data collection to answer questions such as:
•    Does the intervention lead to the desired changes in staff attitudes?
•    Does it lower intent to leave and actual turnover?
•    Does it improve practice and lead to better outcomes for children and families?
•    How long do these changes in outcomes last?
•    Which parts of the intervention have the biggest impact?  
•    Which intervention works best for which types of employees in what contexts?
At the end of the QIC-WD project, we plan to share the efficacy (or not) of eight distinct workforce interventions, their impact on various outcomes and the longevity of the impact. We hope to understand who each intervention works best for, what types of teams each intervention works best for, what contexts each intervention works best in and what components of the interventions are most important in bringing about change.