Article Text

Download PDFPDF

24 100,000 reasons QI works: a vaccine story
  1. Rachel Beitlich,
  2. Amber Larson,
  3. Emily Mishek
  1. Park Nicollet/HealtPartners


Background In late 2020, in the midst of a surge of COVID-19 in Minnesota, the Pfizer/BioNTech vaccine was granted emergency use authorization. HealthPartners CEO Andrea Walsh challenged our organization to provide a plan for administering 10,000 COVID-19 vaccines a day, and Park Nicollet Methodist Hospital leveraged our history of quality improvement to rapidly plan, execute, and scale vaccine administration.


  • Timely vaccination of multiple populations (healthcare workers, high-risk patient populations, and general public)

  • Operate efficiently under several constraints (space, staffing, and vaccine supply)

  • Maximize capacity to fluctuating vaccine allotments

Methods Our work leveraged several QI tools such as value stream mapping (figure 1), weighted average cycle times (figures 2 and 3), visual management (figure 4), and standard work. We utilized the Model for Improvement and sequential PDSA cycles (figure 5):

  • Space required ongoing adjustments in roles, staffing, and flow. We cross-trained to allow for multi-skilled operation, and to accommodate social distancing through the entire process.

  • To reflect our community’s needs, we continuously adjusted to provide a mixed model of first/second dose, delivering service in multiple languages, serving adult and pediatric populations, and ambulatory and non-ambulatory patients.

  • We used predictive modeling to anticipate number of vials needed, standardized drawing process to maximize doses from each vial, and instituted a standby process to eliminate end-of–day waste.

Results In total, during a 6-month operation we administered nearly 100,000 doses.

Abstract 24 Figure 1

Initial value stream map of clinic operations. Scaled to predict staffing needs in order to achieve 10,000 vaccines a day using the same model

Abstract 24 Figure 2

Due to space constraints in post-vaccination waiting room to accommodate social distancing, calculations were needed to understand how many chairs would be needed with different volume/flow through the clinic (and ultimately was the rate limiting step in the total capacity of the clinic)

Abstract 24 Figure 3

Understanding cycle times and staffing needs at different rates of vaccinations per hour. Weighted average cycle times were used to accommodate differing complexity of patients (based on demographics of currently eligible populations). Based on the efficiency of cycle times achieved, different numbers of vaccinators and hours of operation would be needed to achieve volume goals

Abstract 24 Figure 4

Space planning: optimize flow, maximize capacity

Abstract 24 Figure 5

Results, demonstrating rapid cycles of PDSA and frequent adjustments due to changing patient populations, vaccine supply, efficiencies achieved.

Conclusions Our legacy as an organization with a mature culture of improvement contributed to trust of rapid cycles of PDSA, which allowed us to deliver on the Triple Aim:

  • Reducing cost by using existing employees and volunteers

  • Addressing health of the population by targeting multiple patient tiers to optimize supply

  • Delivering experience from invite to injection with sensitivity to patient needs

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.