Optimizing patient choice and care quality: An open-market approach to post-acute transitions

Optimizing patient choice and care quality: An open-market approach to post-acute transitions

  • Marcia Colone

    VP of Care Coordination and Transition Management

    "Aidin leverages untapped capacity in the post-acute provider space. Engaging in this open-market model has increased responsiveness from providers and allows for an even playing field."


As the share of healthcare delivered by 30,000+ post-acute providers continues to grow, hospitals and health systems are working closely with these providers to lower the cost of care and improve outcomes.

Spending on post acute care can often account for more than 50% of an episode’s total cost of care.

Variations in post acute care quality is significant with 30% of all discharges to post acute care being readmitted

Patients are choosing the highest quality post acute providers only 20% of the time

To address this, hospitals and health systems currently employ a number of tactics to influence post acute care providers. These tactics vary in cost and effort but often do not realize their intended goals due to a number of barriers.

There is a need for disruptive innovation in the transitions of care setting as hospitals have to respond to the pressures of the healthcare landscape and better manage post acute care outcomes.


  1. Increase transparency and accountability in post acute provider performance

  2. Empower the patient to choose a provider based on quality, not only on availability/convenience/location

  3. Add efficiency for staff in the post acute referral, choice and reservation process

  4. Improve length of stay outcomes for the hospital


Vanderbilt University Medical Center (VUMC) is a 1,131 bed hospital system with a wholly-owned home health agency and  joint ventures for both inpatient  rehabilitation and home infusion.

For the time period of May 1, 2018 through April 30, 2019, Vanderbilt University Hospital placed 12,495 patients to post acute settings. The dispersion of the referrals to different post acute settings is outlined in the table below:

There is a need for disruptive innovation in the transitions For patients going to one of the above mentioned dispositions, the average length of stay was 9.75 days and the average 30 day readmission rate was 15.1%.

Vanderbilt continues to be challenged by a daily bed capacity problem which creates significant queuing in the ED because floor beds are occupied. Furthermore, the acuity of patients at VUMC coupled with their complex, psychosocial needs creates discharge barriers that are challenging to resolve which delays discharges when the patient is medically ready.of the healthcare landscape and better manage post acute care outcomes.

Solution design

In 2019, VUMC partnered with a leader in care transitions (Aidin) to introduce a technology based solution for post acute referrals. Key solution components highlighted below:

Process re-imagined

Provider database & open market referral

All providers, in network and out of network, indicated which referrals they want to receive. These preferences were managed in a database and continuously updated by Aidin. Staff entered the discharge date, level of care, zip code, insurance information and clinical needs of the patient into the technology platform to query the database for all providers that are possible matches. (i.e. accept the insurance, cover the service area, provide the clinical needs). Based on the list of matches, staff sent the referral to multiple providers and allowed providers to respond with acceptance or declination – this forced an ‘open’ market model for referrals where similar providers were competing to win the patient.

Provider characteristics tracked and displayed

All providers in the database had characteristics or metrics that they were tracked against (subset below). A provider’s performance against the characteristics was visible to all staff members as he/she initiated referrals.

Referral auction step introduced

Once the referral was sent to multiple providers with the ‘best’ characteristics, providers were provided a time limit within which to accept or decline a patient. For each referral, providers earned credit toward their characteristics / metrics. This empowered staff to pick providers with highest quality outcomes, encouraged providers to respond to patients referrals timely and urged  providers to improve their metrics to continue to referrals from VUMC.

Patient/family choice

After providers had submitted their availability, staff printed a choice list for the patient & family that contained information about providers that had accepted the patient. This information packet included pictures, details about the facility and metric performances of those providers (such as Medicare readmission rate). Patients and family were given time to make a decision, after which the staff reserved the bed with the provider using the technology platform.

Implementation method

  • Superusers were selected and  provided training to the technology, the process and the script with patients / providers
  • Computer lab training was then conducted with all social workers and case managers across all hospitals
  • Meetings were conducted with all in network and owned post acute providers introducing them to the technology platform, the new process and scripting for patients.
  • Superusers and staff of the providers were also trained on technology and process
  • After training, a date was decided for ‘go-live’. During go-live a control room was set up to assist staff users as well as provider users on the new process. Daily go-live check in calls were also implemented to ensure no patients are being impacted. Metrics were monitored daily.
  • After ‘go-live’: Staff & providers were provided resources for them to reach out to for technology and operations questions
  • Metrics on outcomes and performance were shared with staff and providers at various forums and difference cadences

Results & conclusions

As seen in Figure 1, 77% of SNF referrals, 72% of home health referrals and 60% of IRF referrals were sent to 2 or more providers.

Figure 1. Volume of Referrals by Disposition & % of Referrals Sent to 2 or More Providers

For every patient, there was an average of 2.5 providers available and willing to provide care – this was even higher for SNF (4.5) and Home Health (3.3) placements (Figure 2). This points to untapped capacitin the post acute provider space that this open marketplace process avails for staff and patients.

Figure 2. Provider Availability per Referral , by Disposition

Of patients given a choice of 2 or more providers, 92% chose the highest quality provider compared to a MedPAC reported avg. of 20% (as seen in Figure 3).

Figure 3. % of Patients Choosing Highest Quality Provider

In this open market process, 89% of patients were placed in the first contact to providers (Figure 4). Additionally, in this new process, only 11% of the referrals required a second touch compared to a 30% baseline that was observed in the traditional process. This elimination of rework contributed to 0.19 day LOS reduction that was observed (Table 1).

Figure 4. % of Patient’s Placed in 1 Contact
Figure 5. % of Referrals Reopened Due to Provider Unavailability
Table 1. Length of Stay Outcomes (Baseline vs Measurement Period)

Lessons learned

  1. When provided with quality metrics, 92% patients are more likely to choose providers that provide highest quality
  2. Engaging in this open market model has increased responsiveness and raised challenges from providers that are owned by VUMC
  3. VUMC “owned” providers are now competing with community providers due to the patient choice process which disallows VUMC to create an exclusive network of our owned providers
  4. The learning curve for staff should be compared to other technology implementations. The transparency of staff engagement is documented due to the time stamps at key intervals of the process
  5. Ongoing education of the organization regarding the process and metrics is critical – it requires focused communication on organization goals such as LOS, among others.


  1. “IRF Quality Reporting Overview.” CMS.gov Centers for Medicare & Medicaid Services, 2 July 2019, https://www.ems.gov/Medieare/Quality-lnitiatives-Patient-Assess ment-lnstruments/lRF-Quality-Reporting/index.html
  2. Tian W. May 2016. An All-Payer View of Hospital Discharge to Postacute Care, 2013 . Healthcare Cost and Utilization Project Statistical Brief #205. Agency for Healthcare Research and Quality.
  3. CMS Program Statistics, 2016. https://www.ems.gov/Researeh-Statisties-Data-and-Systems/Stati sties-Trends-and Reports/CMSProgramStatisties/2016/2016_Utilization.html
  4. MedPAC Report to the Co ngress: Medicare and the Health Care Delivery System, June 2018.
  5. Kothari P and Guzik J. January 2019. Health Care Provider Perspectives on Discharge Planning : From Hospital to Skilled Nursing Facility. New York: United Hospital Fund.
  6. Gadbois E, Tyler D, and Mor V. 2017. Selecting a Skilled Nursing Facility for Postacute Care: Individual and Family Perspectives, Journal of the American Geriatrics Society 65
  7. Mitchell S, Laurens V, and Weigel G. Ma y/June 2018. Care Transitions from Patient and Caregiver Perspectives. Annals of Family Medicine 16(3)
  8. Tyler D, Gadbois E, McHugh J, et al. 2017. Patients Are Not Given Quality of Care Data about Skilled Nursing Facilities When Discharged from Hospitals. Health Affairs 36(8): 1389.