Epic Research is not viewable using Internet Explorer. Please try accessing it with an alternate browser.
Cosmos Study

Shorter Hospital Stays Associated with Patient Portal Use

Abstract: Patients with active patient portal accounts had hospital stays an average of one half to one full day shorter than patients without.
November 17, 2021
Dual-Team Study
Team A:Denise Rasmussen, BSN, RNKieran Gallagher, MPHAnnie Goldsmith, MS
Team B:Phil LindemannEric Barkley

Patients who are more engaged in their healthcare often have better outcomes.1 Because patient portal use is one marker of engagement, we wondered whether access to a patient portal correlated with hospital length of stay for certain conditions. To test this, we looked at two different types of admissions: admissions for new illness (COVID-19) and admissions for chronic illness (heart failure). Patients who were hospitalized for either COVID-19 or heart failure were more likely to have shorter hospital stays if they had an active patient portal account.

Figure 1
Length of Hospital Stays in Patients Who Do and Do Not Use Patient Portals by Age
Length of Hospital Stays in Patients Who Do and Do Not Use Patient Portals by Age
Figure 1: The average length of stay for COVID-19 and heart failure admissions by age group and patient portal status. The solid circles on the left of each bar show the average length of stay for patients with active patient portal accounts, while the hollow circles on the right of each bar show the average length of stay for patients without an active patient portal account.

As shown in Figure 1, having an active patient portal account was associated with a shorter average length of stay in the hospital. The difference was larger for COVID-19 patients than for heart failure patients. In COVID patients, having an active patient portal account was associated with an average length of stay shorter by 0.09 to 1.1 days, while in heart failure patients, using a patient portal was associated with an average length of stay shorter by 0.3 to 0.6 days.

The difference in length of stay was present for all age groups included in the study, though it was smaller for patients 85 years and older. The percentage of patients with an active patient portal account was also smaller in the 85-and-older group than in the other age groups included in the study, as shown in Figure 2 below.

Figure 2
Percentage of Patients with Active Patient Portal Account by Age Group and Reason for Admission
Percentage of Patients with Active Patient Portal Account by Age Group and Reason for Admission
Figure 2: The percentage of COVID-19 and heart failure admitted patients with an active patient portal account during the study period by age group.

While there is a complex interaction of factors that contribute to a patient’s length of stay, such as the patient’s diagnoses, acuity, and social determinants2, our data show that a more engaged patient, as defined by an active patient portal account, may contribute to a decreased length of stay.


These data come from Cosmos, a HIPAA-defined Limited Data Set of more than 120 million patients from 141 Epic organizations including 832 hospitals and 13,421 clinics, serving patients in all 50 states. This study was completed by two teams, composed of clinicians and data scientists, that independently acquired and analyzed data. Both teams were involved in the interpretation of results and drafting of this brief. Overall, the two teams came to similar conclusions.

References

  1. James, J. “Patient Engagement” Health Affairs, February 14, 2013. https://www.healthaffairs.org/do/10.1377/hpb20130214.898775/full/. Accessed September 17, 2021.
  2. Baek H, Cho M, Kim S, Hwang H, Song M, Yoo S. “Analysis of length of hospital stay using electronic health records: a statistical and data mining approach.” PLoS One. 2018;13(4): e0195901.