When we talk about making meaningful improvements in the Health and Human Services (HHS) space, we might assume the only way to do this is with a large budget, greater resources, or highly specialized staff. With these, we know we would have more time, higher performance rates, and greater capacity to give individuals the help they need. However, when larger budgets are not available, we need to look at working more efficiently with what we have.
A prime example comes from sporting data professionals, who use prescriptive analytics to determine future strategies, allocate resources effectively, and mitigate external circumstances. As chronicled in Moneyball: The Art of Winning an Unfair Game, Oakland As General Manager Billy Beane leveraged data analysis to tie for the most wins in Major League Baseball while spending 60% less per win. Like the Oakland As and other sports teams that employ prescriptive analytics, HHS agencies can apply these proven strategies to maximize their existing resources, streamline systems, and elevate service delivery – no hefty budget or large team required.
Gaining an Advantage
Sporting data professionals collect, analyze, and use data to gain a competitive advantage both on and off the field, such as player/team performance and dynamics with machine learning and statistical analysis. Analyzing in-game strategies and non-sporting events identifies correlations with overall team success and creates predictions to help them prepare for future scenarios. For example, when the NFL implemented a rule that allowed the kicking team to start much closer to the receiving team, the Kansas City Chiefs general manager might have wanted to know how the new rule would affect their offensive possessions in the next season. A data professional could create a predictive model to forecast how the rule change affects the potential point gain or loss, which would assist the coaching staff with new game plans.
How this translates to Health and Human Services
Although sports and HHS may seem to be different fields, both can effectively utilize data transformations, predictive models, and quantifiable organizational goals. For HHS, performance metrics might include the number of office SNAP applications processed each day, “one-touch” completion rates, how much time workers spent on cases, and team efficiency rates. If an HHS data professional wanted to analyze the effects of a state government’s new unemployment insurance (UI) requirements, a data professional could create a predictive model to forecast the number of applications and help managers with their staffing plans.
Much like a sporting data professional, an HHS data professional achieves their statistical and analytical objectives through the smart use of statistical methods, visualizations, software, and analytical tools, including:
- Precise, but simple performance metrics with SQL and Exploratory Data Analysis
- Elegant visualizations that effectively communicate outcomes, such as R, Python, and Excel.
- Effective statistical modeling with correlation analysis, regression analysis, ANOVA testing, and additional predictive analysis
How C!A® clients are already benefiting
At C!A, this analytical approach has resulted in real-world outcomes. When the Oklahoma Department of Human Services wished to assess worker performance and the organization’s overall performance, we developed the Oklahoma Performance Dashboard. This dashboard includes metrics derived from Quarterback Rating (QBr), which analyzes quarterback play in the National Football League (NFL). QBr gathers several important Key Performance Indicators (KPIs) and divides them into three different categories (i.e., production, efficiency, and situational), and each KPI is weighted on perceived importance to the end user and combined into one, comprehensive grade.
At C!A, the Performance Score creates HHS worker metrics with approximate one-to-one relationships between the QBr and HHS performance metrics. The following table shows these relationships:
Table 1: Comparison table displaying connections between QBr and HHS KPIs.
Because Performance Score allowed Oklahoma users to view individual KPI grades, they were able to identify performance bottlenecks and target performance deficits. Analyzing worker performance helped them understand the driving factors behind performance changes.
The results were impressive. As can be seen in the following figure, increase in worker attendance and availability led to a three-point increase in the organization’s overall Performance Score between June and July 2024. Notably, administrative staff could further configure the dashboard to display historical performance scores.
Figure 1: Line graph displaying month-to-month change in Organization Performance Score.
In addition to the three-point increase in the organization’s overall Performance Score between June and July 2024 and an additional seven-point increase between July and August 2024, Oklahoma experienced additional performance changes:
| Month | Attendance (5 Points) | Availability (20 Points) | Utilization (20 Points) | Completion Rate (20 Points) |
Transaction Time (10 Points) |
Production (20 Points) | Escalation (20 Points) | Performance Score (20 Points) |
| August 24 | 4.1 | 15.8 | 17.0 | 18.9 | 7.3 | 17.0 | 5.0 | 85.1 |
| July 24 | 3.7 | 9.8 | 17.1 | 18.8 | 7.2 | 16.9 | 5.0 | 78.5 |
| June 24 | 4.0 | 6.1 | 16.7 | 18.8 | 7.1 | 17.7 | 5.0 | 75.4 |
Table 2: Rating table displaying monthly Performance Scores broken out by KPI.
Conclusion
Th HHS data professionals, much like their sporting data professional counterparts, break down their output into simple volume and worker performance metrics, they can predict their outcomes as well as the quality of their communities and worker capacity. By adapting analytical and statistical techniques from the sports world in the HHS sector, we can foster tremendous and quantifiable benefits for the people we serve.
Recent Comments