Hospital leaders across the country are watching their subsidization of shift-based specialties, such as hospital medicine, emergency medicine, and critical care, steadily rise as provider compensation continues to increase 2-3% annually and reimbursement stays flat at best.[1] In recent years, investment per physician FTE has surpassed $200,000 for shift-based specialties.[2] Many are asking whether they have the most efficient provider coverage models for their inpatient units, ICUs, and EDs. Those proactively taking a more sophisticated approach to analyzing and managing their models, and engaging physicians in efficiency improvement initiatives through compensation design, are reigning in this investment.
Optimizing efficiency of provider time is a key objective for any specialty, however, shift-based specialties present a very different challenge: namely, patients are not scheduled. Some sources of patient volume have a degree of predictability, such as inpatient stays following scheduled surgeries or seasonality in certain injuries and illnesses, but other sources do not, resulting in unpredictable variations in total patient load and arrival patterns. So, the question becomes, how do we ensure we are efficiently scheduling provider capacity to meet patient demand? And the most effective organizations also ask, how can we reduce the number of providers we need to cover the same patient demand?
Ultimately, the goal is to maximize direct patient care time (or, billable time) per unit of provider scheduled time and eliminate time in which they are not (i.e., keep providers at top of licensure). The remainder of this article will summarize a four-step process to analyzing and managing shift model efficiency, and then discuss alignment with compensation design.
Analyzing and Managing Shift Model Efficiency
Step 1 – Benchmark ‘Per Provider’ Theoretical Capacity
A critical first step in answering these questions is identifying the theoretical, or ideal, capacity of an individual provider per unit of time (e.g., patients cared for per day by a hospitalist or patients treated per hour by an emergency medicine physician). Specialty societies, such as the Society of Hospital Medicine (SHM) and American College of Emergency Physicians (ACEP), commonly report on provider throughput benchmarks[3],[4], which can be used as guidance. These benchmarks typically will have a range, as theoretical capacity is influenced by many factors, such as average patient acuity, service commitments (e.g., hospital designations, academic programs), and geography. Another consideration of increasing importance is the prevalence of value-based reimbursement in a hospital’s market. Not all revenue is driven by volume of patients, and performing under those models may require more effort (time) per patient. It is important to evaluate a hospital’s circumstances relative to these factors to determine the most relevant benchmark to peg theoretical capacity.
Step 2 – Define ‘Per Provider’ Practical Capacity
Commonly, hospitals will discover that actual average performance is less than the theoretical. This is due to an array of hospital-specific inefficiencies, including but not limited to suboptimal provider scheduling, high average patient length of stay, poor access to and/or collaboration with support staff, long turnaround times from imaging and lab, limited access to specialty consults, and physical constraints (e.g., bed availability, facility design, etc.).
The next step is to determine practical capacity, which is defined as maximum performance under the existing conditions. A deeper dive into the data, such as granular analysis by day of week, portion of the day, or season of the year, will reveal a range of performance. Identifying a point at the higher end of that range that occurs at some frequency (i.e., don’t pick outliers) indicates what providers could practically achieve more typically. This will become an initial target of performance with improved scheduling efficiency. We will readdress the gap between theoretical and practical in step 4.
Step 3 – Develop a Provider Staffing Strategy to Improve Demand-Capacity Alignment
Provider staffing strategy should reflect thoughtful analysis of patterns in patient demand. Hospitals should strive to manage staffing to practical capacity, rather than historical practices. Very simply, dividing patient load by per provider practical capacity equals the required scheduled providers. However, as mentioned earlier, there is a lot of variation in patient load making it challenging to predict the ideal staffing model. Some variation can be eliminated by meaningful granularity in the schedules developed. For example, a hospitalist model often benefits from seasonal schedules to account for seasonal trends (e.g., flu). As another example, Saturday through Monday in an ED is very commonly busier than Tuesday through Friday and may warrant different provider schedules. EDs also benefit from analysis of when shifts begin throughout the day to match hourly arrival patterns.
Many hospitals will design shift models around either historical peak or average patient load, but these hospitals will sacrifice either financial sustainability or quality of care, respectively. Effective hospitals will seek pockets of latent provider capacity and innovate staffing models to eliminate these occurrences. As an example, flexible models are a more contemporary approach, which include shift types that allow provider staffing to adapt daily capacity to patient demand. Under these models, a base level of coverage may be set at average, or slightly above average, patient load and additional shifts are established with the providers’ understanding that they may be called in or sent home early depending on actual patient load. These models also include triggers for reassessment of the model when patient load per provider frequently exceeds practical capacity. A downside to flexible models tends to be provider satisfaction and burnout, thus it is critical to strike a balance of base coverage and flex shifts that does not overly burden the providers or result in them constantly functioning beyond their capacity. In 2021, 47% of physicians reported feeling burned out, with common attributions to being pushed to work beyond their physical limits and reduced staff [5].
Another area that analysis can help optimize staffing models is with regards to provider mix. Depending on state laws around APP scope of practice, some hospitals have the opportunity to deploy more cost-effective care models. Further granularity of patient load by acuity can identify opportunities to schedule APP shifts in place of physicians. In 2020, 83% of organizations report utilizing APP, up from 51% in 2012.[6] Additionally, APPs rather than physicians can sometimes support hospital patient access initiatives, such as expansion of operational hours and/or sites of care (e.g., micro/neighborhood hospitals, freestanding emergency rooms, night and weekend coverage, and advanced home health and telemedicine coverage). These shifts tend to be low volume, low acuity single coverage that an APP can handle with a physician available on-call.
Steps 2 and 3 should be performed regularly to ensure that scheduling practices continue to reflect practical capacity and shifts in patient demand patterns most appropriately.
Step 4 – Closing the Gap Between Practical and Theoretical Capacity (iterative)
Steps 1 through 3 are a great initial approach to achieving a more efficient staffing model for shift-based specialties. However, a gap between practical and theoretical capacity implies lingering latent provider capacity, and as such, opportunities to further improve staffing models. This gap is a cost to the hospital as it means that more providers are needed for the coverage than benchmarks would indicate, and the wider the gap the greater the cost. Hospitals with the requisite resources may also adopt an iterative process of optimizing provider capacity.
Continuous assessment of throughput and system constraints should reveal root causes behind the gap. Some causes, such as facility limitations, may be too expensive to resolve, at least in the near term. However, factors that lead to longer than ideal patient length of stay can commonly be addressed and are often the largest contributors to depressed provider capacity. Each time an improvement is implemented, provider practical capacity will increase and provider need will decrease.
Drawing upon core operations management concepts, such as those described in Dr. Eliyahu Goldratt’s Theory of Constraints[7], achieving theoretical capacity requires a systematic process of identifying the constraint to throughput and focusing improvement efforts on that constraint. Constraints tend to be related to hospital policies, paradigms of thinking around care or operations, or the physical (i.e., utilization of support staff, management of space, availability of equipment). Hospitals that constantly brainstorm questions to challenge current practice and processes, as well as innovations to test, will enable higher productivity from their providers.
Alignment with Compensation Design
Continuous improvement of coverage model efficiency is a multi-faceted challenge and absolutely requires collaboration and engagement from your provider partners. Compensation or group funding design is an effective vehicle for creating this essential alignment.
Compensation and funding methodologies for shift-based specialties continue to evolve across the industry to better reflect the full scope of services and contributions by providers to programmatic improvements, such as those related to efficiency. Hospitals are implementing hybrid approaches, which provide stability through a baseline guarantee and upside potential through incremental incentives for production, quality, and efficiency. Non-productivity variable compensation for shift-based specialties has increased to over 8% in recent years, implying available compensation at 10% or more, and ranging up to 20-30%.[8]
Below are a set of questions to consider in assessing how well your provider compensation or funding design aligns with your shift-model efficiency initiatives:
- Are productivity targets aligned with relevant benchmarks, as discussed above?
- Are productivity and quality incentives appropriately balanced relative to organizational focus and reimbursement mix?
- How are excess or flex shifts incentivized to minimize need for FTE and/or locum tenens?
- Are there appropriate paid provider leadership positions related to efficiency, such as department or hospital performance improvement committees?
- Is citizenship (e.g., participation in performance improvement initiatives, daily communication and huddles with nursing) a base expectation, or is it recognized through incentives?
- How do benefits (PTO, retirement contributions, wellness) support balance?
Providers are an essential voice in improving shift model efficiency and can help expedite transformational changes. However, their engagement takes time and has an opportunity cost, therefore it is critical that compensation and funding design supports strong collaboration.
Authors: Sean Cappello and Jessica Reed
Contributions: Clark Bosslet and Kelly McFadden
[1] ECG Physician and APP Compensation Survey Reports
[2] ECG Medical Group Cost Survey Reports
[3] https://www.hospitalmedicine.org/practice-management/shms-state-of-hospital-medicine/
[4] https://www.acepnow.com/article/staffing-ed-appropriately-efficiently/
[5] Leslie Kane, “Physician Burnout & Depression Report 2022: Stress, Anxiety, and Anger.” Medscape. January 21, 2022. https://www.medscape.com/slideshow/2022-lifestyle-burnout-6014664
[6] SHM SoHM Reports, nonacademic hospital medicine groups serving adults only
[7] https://www.tocinstitute.org/theory-of-constraints.html; https://www.leanproduction.com/theory-of-constraints/
[8] ECG Physician and APP Compensation Survey Reports
