By Mike Hill, JD, MSHA, director of practice operations, Rush University Medical Group; Erin Provagna, MHSA, administrative fellow, Rush University Medical Center; Shannon Driscoll, MHA, associate vice president, practice operations, Rush University Medical Group; and Sara Turley, MBA, FACHE, chief of staff, Rush University Medical Center.
As ambulatory volume increases, it is essential for clinical practices to utilize efficient staffing models that align with patient demand. Optimal, appropriate full-time-equivalent (FTE) staffing is contingent upon the right staff performing the right tasks at the right time.
National staffing benchmarks exist, including those published by MGMA; however, assessing and operationalizing a consistent staffing model across many clinics can prove challenging for a large medical group with multiple practice variations and levels of efficiency.
Use of a staffing-to-demand (S2D) model allows comparison of an individual practice’s current staffing levels to what is required to meet the specific patient demand of the practice. S2D combines two key elements — standard work and staffing models — to ensure ambulatory clinics are appropriately staffed to provide high-quality, reliable care to patients, as well as to ensure each staff member is functioning at the top of his/her license. The result of conducting an S2D analysis is a highly specific ambulatory staffing model, which can be further leveraged across a medical group to share best practices and drive operational success.
The S2D model
Rush University Medical Group (RUMG), which consists of more than 100 clinical practices throughout Chicagoland, lacked a standard and data-driven approach to clinical staffing, leading to considerable variation among practices. While ambulatory clinics inherently have some variation in the tasks performed before, during and after an office visit, these differences can lead to unnecessary delays and have negative implications on operations and finances.
For clinical practice staff at RUMG, these variations included how long it took staff to perform similar tasks, such as rooming a patient or coordinating care. Additionally, the type of staff utilized for routine tasks varied by practice. For example, rooming and assessing patient vitals, which is standard work for a clinic visit, was performed by a variety of different job titles across the medical group. Further, RUMG noticed that tasks associated with care coordination (including answering patient messages, phone triage and completing prior authorizations) were increasing year over year. This often resulted in practice leaders requesting additional staff based on perception rather than actual data.
In response to this, S2D was implemented across all RUMG practices, illuminating practice-level staffing needs. RUMG utilized MGMA staffing benchmarks as an essential starting point and then conducted S2D analyses to discover variances between current and needed staffing at the practice level. MGMA staffing models provided a guide as to how other national practices utilize staffing per visit and staffing per physician clinical FTE. If the internal S2D analysis did not align within a deviation of the MGMA models, the medical group sought to review the S2D output.
The success of any new project requires a robust change management strategy, and S2D was no different. The medical group communicated to senior leaders, directors, managers and frontline staff regarding the importance and benefits of the S2D work for each practice. Numerous conversations occurred between the project team and leadership as to the application of the model to each leader’s unique practice. Additionally, the medical group engaged and explained the work to the physician leaders. Ensuring all practice managers were aware of the project prior to observers entering the clinic to begin the S2D analysis was crucial to the project’s success.
Calculating current cycle time
As a first step in measuring standard work, a team conducted in-clinic observations of clinic coordinators, medical assistants (MAs) and registered nurses (RNs). The observation team documented standard tasks and the time it took to complete each task, as illustrated in Figures 1, 2 and 3.
The number of observations per task varied by practice and staff size, with the goal of capturing an average or typical task duration. For example, a smaller practice with fewer MAs may only require one or two observations to determine the average task duration. While this assessment has focused on work performed by clinic coordinators, MAs and RNs, it can easily be applied to any role in the practice.
Calculating needed capacity
Utilizing the cycle time for each task, and considering the practice’s current and budgeted visit volume, RUMG took a data-driven approach to compute necessary staffing volumes. Figure 4 illustrates the standard tasks performed by the MAs and incorporates budgeted volume, which is usually patient volume or can be specific to the task (e.g., the number of messages or procedural volumes). Next, the total number of hours dedicated to each task is calculated, and then converted into an FTE by dividing by 2,080. An additional 0.20 FTE was added to account for any non-productive time, such as vacation.
An important consideration in calculating needed capacity is quantifying care coordination. RUMG struggled with quantifying the time needed to complete some types of work, such as phone messages, prior authorizations or other tasks related to care coordination. In collaboration with information systems, RUMG was able to request reports that accurately captured time stamps for work performed in the EHR, which allowed for more accurate data.
Once current and needed capacity were calculated, RUMG created a dashboard for each clinic, which compared actual versus needed FTEs, as shown in Figure 5. The variance for each clinic, positive or negative, was made available to practice and medical group leadership, and provided key management insights. These dashboards have been utilized for budgeting and hiring of staff, as well as for insight as to which clinics can share staff across multiple practices. As a practice’s visit volume increases or decreases over time, this data informs requests for additional positions.
Historically, the medical group had a position control committee that reviewed requests for incremental and new positions for staff. Most of the feedback regarding the committee were that asks were made instinctually rather than through a quantifiable presentation. The committee evolved to require a quantifiable number based on the S2D framework. All positions — either replacement, incremental or new — are brought to the position control committee and reviewed based on the S2D dashboard.
The previously private practices in the medical group had numerous variations in operations before joining the unified physician group. A huge benefit of S2D was the creation of standard work across all clinics. The standard work was developed through best practices identified by the observers. For example, if Clinic A is rooming patients within nine minutes, and Clinic B is rooming patients within six minutes, then the variance leads the team to explore what Clinic B is doing to be more efficient. Leadership has since standardized the best practice of Clinic B and applied it across the medical group. There were also inefficiencies identified throughout the practices that were eliminated. Ultimately, the more efficient a clinic becomes, the less time it would take to complete a particular task; e.g., a decrease in cycle time across the practice.
Top of license work
Ambulatory practices are most effective when staff operate at the top of their license, doing the tasks they were hired to do. If not, we need to shift the right work to the right people. The observations throughout all clinics that led to the S2D model illuminated the medical group’s focus on top of license work. For example, the observers found in many practices that nurses, and even sometimes physicians, were rooming patients. As a result, the medical group was able to apply a consistent model of MAs rooming the patients.
The S2D model is dynamic and should be evaluated by practice leadership periodically. The medical group has asked that each practice refresh the S2D model at least quarterly. A next step in gauging the overall success of the S2D model will be to analyze the impact on patient, provider and staff engagement. If clinics have the right staff performing the right tasks, we would expect to see an increase in quality care for patients and highly engaged employees. Of course, no quantifiable model can take the place of empowering the leadership team and staff to improve operational efficiency.
Acknowledgment: The authors would like to thank and acknowledge the Rush University Medical Group leadership team, including Richa Gupta, Anisa Jivani and Laura Kayler, for their work in constructing the staffing-to-demand playbook, which led to this article.