Skip To Navigation Skip To Content Skip To Footer
    Insight Article
    Home > Articles > Article

    Click here after reading to take the ACMPE article assessment.

    By Devin McConnell, capacity management analyst/project leader, UW Medicine | Valley Medical Center; Rolanda Parker-Saechao, capacity management analyst/project leader, UW Medicine | Valley Medical Center; and Jose Mari G. Lansang, BSN, RN, ambulatory quality manager/project consultant, UW Medicine | Valley Medical Center.

    Healthcare organizations strive to strike a balance between provider workload and patient volume to set panel size benchmarks, which are usually derived from supply and demand calculation and results between 1,800 and 2,500 patients.1,2,3,4 One group even incorporated new technology such as machine learning to account for variables of time across different care encounters.5

    Many healthcare organizations have adopted some form of risk stratification to help improve awareness of the additional time needed beyond the in-clinic visit to manage the care of patients appropriately.6,7 Clinical pathways and clinical practice guidelines affect demand by setting the frequency for certain levels of management, from a yearly follow-up to every three months.

    Lastly, there is the ever-present fiscal element: balancing revenue with growth. A University of Wisconsin study helped define the disparity between the time associated and the work relative value unit (wRVU) with an encounter type. The study identified that the workload needed for non-face-to-face activities ranges from 32% to 63% for Level 3 E/M of an established patient (code 99213).8

    Most of the methodologies focus on availability, whether by the count of bookable appointment slots or time needed, and some have done a reasonable job of accounting for non-face-to-face care events. However, there is not a widely accepted measure of quantifying the primary care provider’s (PCP) overall workload, which accounts for the time needed to accomplish in-clinic encounters as well as non-face-to-face patient care activities.9

    Incorporating the Quadruple Aim

    This work presents a comprehensive logical tool that incorporates patient access needs, sparks conversations about quality of care, acknowledges the time needed for care management and addresses financial fitness. The Patient Capacity Management Tool, developed by Valley Medical Center (VMC) Clinic Network, is in a spreadsheet platform but can be incorporated into a business intelligence platform, which would offer the added enhancements of dashboards and drill-down capabilities. This tool addresses the four primary goals of clinic operation: optimizing schedules, improving access through panel size adjustment, panel management and meeting productivity goals.

    Schedule optimization

    This tool is designed with an emphasis on work-life balance for the healthcare provider; therefore, it accounts for planned time away from the clinic for personal and professional development. The formula uses a ratio of 44 patient contact weeks for every 52 employed weeks (44/52). This ratio, 85% of the full-time equivalent (FTE), provides an estimated number of net bookable clinic days. The right side of Figure 1 shows an example of a provider’s actual slots and time in the clinic, which matches the necessary time and slots calculated for the FTE, shown on the left side. In this example, the provider is meeting the required minimum patient contact hours of 28.8.


    • Patient contact hours: Time allocated to see patients every week, and providers are assigned with 36 hours to see patients in a 40-hour workweek.
    • Mixed model ACP: Advanced care practitioners (ACPs) whose clinical time/contact hours are split between seeing patients and providing administrative support to other providers.
    • Base visit duration: The unit time per visit slot, which can be a 15-minute slot or a 20-minute slot, depending on the clinic’s preference.
    • Patient hours: Dedicated time for face-to-face visits.
    • Focus hours: Dedicated time for non-face-to-face tasks, such as care coordination and documentation.
    • Team adm cover hours: Amount of time needed by mixed-model ACPs to provide administrative support to other providers.
    • 95% panel size range: The lowest and maximum panel size estimates were reduced by 5% to provide a buffer.
    • Weighted value: The percentage value that needs to be applied to decrease the panel size according to the number of high-risk patients.

    By reviewing past encounter data, operational leaders can compare an individual provider’s occurrence of team visits with the clinic average to drive decisions for adjusting schedules. Team visits happen when a provider sees another’s patient due to lack of access or the provider is out of office. In contrast, team administration cover means that the provider’s employment includes providing medical administrative support to the team, such as medication refills and completing prior authorization. In this case, the provider is not tasked to do team administration cover as part of his or her employment. However, the provider can see other teams’ paneled patients if there is an access issue. The impact of this access problem will be discussed further in the Patient Capacity Management Tool.

    Improving access through panel size adjustment

    The next section of the Patient Capacity Management Tool is designed to account for patient complexity using VMC Clinic Network risk stratification scores, which are composed of scores from the modified LACE Tool10 and in-house scoring system of patient medical history, insurance status, medication intake and continuity of care.

    The Capacity Management team applied weighting values based on the percentage of high-risk patients in the provider panel to account for the additional time involved and to generate a range of panel sizes (see Figure 2). The tool estimates a maximum panel size by dividing the number of projected visit slots with the average number of visits per year (AVPY) from the past two years of visits.

    For this provider, the projected encounter is 3,036 divided by 2.33 AVPY, which provides an absolute maximum panel size of 1,304; the 95% maximum panel size is 1,239. The maximum panel size does not account for the number of complex patients present in the current panel size, and this number is primarily based on historical data and available visit slots. The AVPY is a variable dependent upon each patient’s tendency to be seen in the clinic, and the value increases or decreases as patients visit their provider more frequently or less often.

    On the other hand, the minimum panel size calculation takes account of patients’ complexity via the weighted value. For this example data, the weighted value (25%) is determined by finding the percentage of patients in the current panel with a high-risk score, which is 19.9% (See Figures 2 and 3). Once the weighted value is established, the minimum panel size (913) is determined by multiplying the maximum panel size with the difference between 100% minus the weighted value and 5% buffer [1304 (1 - 0.25 – 0.05)]. The percent full of median calculation is intended to inject the “human factor” into this data-centric tool by encouraging conversation between providers and clinic leadership. In this example, the current panel is 36% above the median, with 19% of the current panel considered high-risk. Both stakeholders can discuss whether to move toward the lower end or upper end of the panel size range.

    Panel management

    There are several details about a provider’s panel in this section of the tool, such as panel count, panel status and template utilization (see Figure 3). In the lower-left corner, there is a table for tracking and trending of organizational key performance indicators (KPIs). This information allows leaders to understand what (if any) changes may be needed to the schedule template or if there is a need to leverage other resources to support the provider. The right side of Figure 3 shows data from the clinic’s other providers. In this example, the panel status is a soft close, and it is likely appropriate since the productivity exceeds the MGMA benchmark target, and the abovementioned data also support this conclusion.

    To improve panel management further, template slot utilization should be reviewed. In this case, the low utilization of same-day slots and overutilization of unblocked slots indicate that it may be necessary to revert some of the same-day blocks to unblocked slots. Having more unblocked slots may improve the scheduling lag for new patients, which is above the organization’s threshold of 10 days. Furthermore, adjustments of the template may improve access for the current paneled patients since other providers have seen 9% of paneled patients.

    Another aspect that should be considered is the correlation between the percentage of high-risk patients and wRVU per encounter. Compared with the rest of the team, it may indicate a need to review coding practices. In this example, the provider is far above the clinic average of high-risk patients per panel (10.9%). On the other hand, the provider is just 0.3 points above the clinic average of wRVU per encounter. In instances of a wider gap, a chart review of high-risk patients may help uncover the coding practice of the provider and may help prevent the provider from billing a lower-complexity visit for a complicated one. Additionally, a review of the visit frequency of patients with chronic disease will indicate if it is aligned with the current clinical pathway.

    Meeting productivity goals

    Taking account of the provider’s average wRVU per encounter allows this Patient Capacity Management Tool to identify potential impacts on productivity as well as assist clinic leaders to discuss goals. Figure 4 shows the total number of completed encounters year to date (YTD). It is used to determine the current average number of visits per month, which helps provide an estimate of the total encounters for the current fiscal year. Moreover, this section of the tool provides a gauge of whether a provider will meet the MGMA benchmark target (65th percentile) by calculating the number of encounters seen and the provider’s average wRVU per encounter. The provider in this example is projected to go beyond the 65th percentile goal (2,511 encounters) based on the 1,822 YTD encounters and applying the average to the four remaining fiscal year months. Conducting this review quarterly helps adjust the provider’s schedule proactively, especially when onboarding new providers.


    The Patient Capacity Management Tool was piloted with 20 PCPs in two clinics for three months. There was a 15% improvement in the average number of days for scheduling lag for new patients, and 41% improvement for established-patient scheduling lag (see Figures 5 and 6).

    The balanced measure of no-show and same-day cancelation rate also improved by 20% for both pilot clinics. On the contrary, there was no significant change in the schedule utilization rate, which stayed around 95%.

    Overall, provider feedback was positive: Of the 20 providers, only one expressed dissatisfaction, and only five voiced concern about the recommended change, while the rest fully embraced the tool.

    These positive results, paired with the ability to use a standardized, objective tool to assess supply and demand across the Clinic Network, led senior leadership to move forward with the tool in the remaining 11 primary care clinics. There is a plan-do-study-act (PDSA) culture in the organization, and the leadership understood that the tool would be adjusted as more information and feedback is gathered across the lifetime of the tool. Lastly, having a standardized tool can continue the movement toward the centralized management of the provider schedule.


    The creation of a standardized tool is dependent on the organization’s needs, bandwidth and data availability. Having a well-established foundation should not be a barrier to taking a small step to becoming a highly reliable organization. Including the stratification scores in the tool can help visualize the panel complexity in addition to provider verbal reports.

    The team plans to monitor data to help validate the tool from sources such as the organization’s internal Annual Employee Satisfaction Survey, third-next-available appointment report, Clinician & Group CAHPS Survey, and the Agency for Healthcare Research and Quality (AHRQ) Surveys on Patient Safety Culture. These data sources can help provide an overall perception of staff and providers on the workload and support from their leadership. Furthermore, these data sources can help provide feedback from the patients regarding their access to care. The team will continue to do more analysis on the relationship between KPIs, such as the stratification scores versus wRVU, to help provide a more comprehensive way to improve the panel management and financial health of a clinic.

    Among others, measuring, analyzing feedback, opening to evidence-based practice and continuous oversight of a patient capacity management initiative can help improve patient access needs, kick-start the conversations about quality of care, create the time needed for care management and sustain financial fitness.


    1. Margolius D, Bodenheimer T. “Transforming primary care: from past practice to the practice of the future.” Health Affairs. 2010 May;29(5):779-784. doi: 10.1377/hlthaff.2010.0045. PMID: 20439861.
    2. Raffoul M, Moore M, Kamerow D, Bazemore A. “A Primary Care Panel Size of 2500 Is neither Accurate nor Reasonable.” J Am Board Fam Med. 2016 Jul-Aug;29(4):496-499. doi: 10.3122/jabfm.2016.04.150317. PMID: 27390381.
    3. Rossi M, Balasubramanian H. “Panel Size, Office Visits, and Care Coordination Events: A New Workload Estimation Methodology Based on Patient Longitudinal Event Histories.” MDM Policy & Practice. 2018 May. doi: 10.1177/2381468318787188.
    4. Murray M, Davies M, Boushon B. “Panel size: how many patients can one doctor manage?” Fam Pract Manag. 2007 Apr;14(4):44-51. PMID: 17458336.
    5. Rajkomar A, Yim JW, Grumbach K, Parekh A. “Weighting Primary Care Patient Panel Size: A Novel Electronic Health Record-Derived Measure Using Machine Learning.” JMIR Med Inform. 2016 Oct 14;4(4):e29. doi: 10.2196/medinform.6530. PMID: 27742603; PMCID: PMC5086026.
    6. Hartley W, Horton F, Cuddeback J, Stempniewicz R, Stempniewicz N. “Why Panel Size Matters: Operational considerations and risk adjustment.” AMGA. 2018 June. Available from:
    7. Kamnetz S, Trowbridge E, Lochner J, Koslov S, Pandhi N. “A Simple Framework for Weighting Panels Across Primary Care Disciplines: Findings From a Large US Multidisciplinary Group Practice.” Qual Manag Health Care. 2018 Oct/Dec;27(4):185-190. doi: 10.1097/QMH.0000000000000190. PMID: 30260924; PMCID: PMC6166700.
    8. Arndt B, Tuan WJ, White J, Schumacher J. “Panel workload assessment in US primary care: accounting for non-face-to-face panel management activities.” J Am Board Fam Med. 2014 Jul-Aug;27(4):530-537. doi: 10.3122/jabfm.2014.04.130236. PMID: 25002007.
    9. Ibid.
    10. Learn more about the LACE index and risk score calculations at

    Explore Related Content

    More Insight Articles

    Ask MGMA
    Reload 🗙