Results Summary
What was the research about?
Patient portals are secure websites where patients can view their health records, view test results, send messages to their doctor, and ask for prescription refills. Patients with chronic, or long-term, health problems such as asthma, diabetes, or heart disease must often coordinate their care across different doctors in multiple locations. Patient portals may be especially helpful for these patients.
This study focused on a patient portal that a large health system in California uses. The research team wanted to learn
- What are the characteristics of patients with chronic health problems who use or don’t use the portal?
- Why do patients use—or why do they not use—the portal?
- How does patients’ use of the portal relate to their use of healthcare services?
What were the results?
The research team found that patients with chronic health problems who were younger or white were more likely than others to use the patient portal. Among patients who said they didn’t use the portal, the most common reasons were
- Wanting to get care in person or by phone (54 percent)
- Not often using a computer with access to the internet (41 percent)
- Being unsure of what was available on the portal (15 percent)
Patients who used the portal said it
- Helped them get health information (92 percent)
- Helped them manage their health care (92 percent)
- Was convenient (90 percent)
Compared to portal users who didn’t report any of these benefits, those who reported benefits were more likely to also report that using the portal improved their health.
Patients who used the portal had more doctor office visits, fewer emergency room visits, and fewer preventable hospital visits than patients who didn’t use the portal.
Who was in the study?
The study included adults with chronic health problems. In the first part of the study, the research team surveyed 1,824 patients. Of these patients, 56 percent were white, 13 percent were Asian, 10 percent were black, and 10 percent were Hispanic. About 12 percent of patients were 18–44 years old, and the rest were older.
In the second part of the study, the team reviewed patient portal and health records from 273,815 patients. Of these patients, 54 percent were white, 18 percent were Asian, 17 percent were Hispanic, 9 percent were black. About 43 percent of patients were 18–64 years old, and the rest were older.
What did the research team do?
The research team worked with patients, doctors, and other health system staff to design a patient survey. The survey asked patients about the reasons they did or didn’t use the portal. The team mailed the survey to patients in the health system and analyzed the results.
The research team also looked at the patient portal and health records of patients in the health system. Using these records, the team saw which patients used the portal. The team also compared the use of healthcare services by patients who did and didn’t use the portal.
What were the limits of the study?
The study took place in a single health system in California with an established, free patient portal. Results may be different in other health systems. Also, results may be different for patients who must pay to use a portal. Results may also differ for patients who don’t have chronic health problems. Finally, the research team can’t say for sure if patients’ use of the portal caused the differences in how patients used healthcare services.
Future research could look at patient portal use in different places or with different patients. Researchers could also explore ways to reduce barriers to using portals.
How can people use the results?
Patients can use the results to learn more about how using a patient portal might affect healthcare experiences. People who run healthcare systems could use the results to help decide whether to provide portals for patients with chronic health problems.
Professional Abstract
Objective
To examine (1) which patients with chronic health conditions do and do not use patient portals and why, and (2) how portal use correlates with healthcare experiences for patients with chronic conditions
Study Design
| Design Element | Description |
|---|---|
| Design |
|
| Population |
|
| Interventions/ Comparators |
Not applicable |
| Outcomes |
|
| Timeframe |
|
Patient portals are secure websites where patients can access their health records. In this study, researchers, patients, clinicians, and other health system staff collaborated to design a survey to understand the drivers of and barriers to portal use by patients with chronic health conditions. The survey also collected information about characteristics of portal users.
The study took place in a health system in California that implemented a patient portal with free access for patients in 2005. The portal contains patients’ medical information and interactive tools to assist patients in participating in their health care. The researchers sent surveys to randomly selected patients from the health system’s registry of adults with chronic conditions, oversampling patients with multiple chronic conditions. A total of 1,824 patients completed the survey. Of these, 56% were white, 13% were Asian, 10% were black, and 10% were Hispanic. About 12% were 18–44 years old, which researchers defined as younger in significance testing; the rest were 45 and older.
The researchers also conducted a retrospective observational cohort study of automated data from 273,815 patients with chronic conditions, examining associations between clinical events and characteristics of portal users and nonusers. Of these patients, 54% were white, 18% were Asian, 17% were Hispanic, and 9% were black. About 43% were 18–64 years old, which researchers defined as younger in significance testing; the rest were 65 and older. The automated data came from measures of portal use, health records, and other automated health system databases.
Results
Characteristics of patient portal users. Both the survey and automated data indicated that patients who were younger or white were more likely to use the portal (p<0.05). Survey results showed associations between portal use and at least some college education, household income above $40,000, access to a computer and mobile device, and regular internet use (p<0.05). The automated data indicated associations between portal use and a more complex clinical condition, more clinical events, copayments for in-person visits, and higher engagement in health care (p<0.05).
Barriers to patient portal use. Among survey respondents who did not use the portal, the most common reasons reported were preferring to get care in person or by phone (54%), not regularly using a computer with an internet connection (41%), and being unsure of what information was available on the portal (15%).
Experiences of portal users. Among survey respondents who used the portal, 90% reported using it for convenience, 92% reported it was useful for data and information management, and 92% reported using the portal to manage their health care. Compared with portal users who did not report any of these benefits, those who did were more likely to also report that using the portal improved their health (p<0.05).
Clinical events. Portal users had more office visits, fewer emergency department visits, and fewer preventable hospitalizations (p<0.05).
Limitations
The study took place in a single health system with a well-established patient portal that was free of charge. Results might be different in other health systems or for patients who must pay for portal use. Because the study focused on patients with chronic conditions, the results may not be generalizable to other patients. Finally, because the study used a cross-sectional survey and observational automated data analyses, the results cannot establish that portal use caused the observed differences in outcomes.
Conclusions and Relevance
Patient portal use varies by user characteristics, technology access, and clinical characteristics. Portal users had more doctor office visits, fewer emergency department visits, and fewer preventable hospitalizations. These findings suggest that portal use may increase engagement and reduce health events that lead to emergency and hospital care.
Future Research Needs
Future research could explore patient portal use in different settings or with different patients. Researchers could also examine methods to reduce barriers for portal nonusers.
Final Research Report
View this project's final research report.
Journal Citations
Article Highlight: Access to a patient portal can increase engagement in outpatient visits by patients with diabetes and those with multiple complex chronic conditions, according to a PCORI-funded study spotlighted recently in PLOS One. The study showed that portal use was associated with significantly fewer emergency room visits and preventable hospital stays for patients with multiple complex conditions. By increasing patient office visits, a portal could potentially help clinicians address unmet clinical needs and reduce health events that lead to emergency and hospital care. The observational study compared visit rates for 165,000 patients with and without portal access in a large healthcare system that implemented a patient portal.
Results of This Project
Related Journal Citations
Peer-Review Summary
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented, and the researchers made changes or provided responses. Those comments and responses included the following:
- The reviewers were concerned that the report did not adequately acknowledge a significant limitation of the study design that could lead to systematic bias related to who did and did not access the web portal. The researchers rebutted that they recognized this problem but because they were aiming for a real-world test of the portal, they accepted this trade-off and used the best possible statistical approaches to account for its effects. They also added a limitation in their discussion of the study, noting that there may be unmeasured patient factors that could account for use or nonuse of the web portal and also influence the study outcomes.
- The reviewers requested more justification for the use of subsets of patients with specific health concerns in some of the analyses of web portal use. The researchers explained that their potential population was very large, and so they needed to use subgroups for some of the usage analyses. They chose specific subgroups for some of the analyses to illustrate how use of the web portal would affect the communication between physicians and patients within a specific chronic disease group.
- The reviewers questioned why the authors included only a limited number of indicators of disease severity in the measures for this study, when many more were likely available in the rich dataset. The researchers responded that they chose measures that were readily available and felt that the measures they chose were good indicators of disease severity and are captured precisely in time. They did add a limitation to the Discussion section acknowledging that there were other indicators of disease severity that they could have examined.