State of the research on online patient portals

By Courtney Lyles, PhD

My main area of research over the past few years has focused on getting diverse patient groups to engage in accessing their medical record information online, through websites called online patient portals that are linked to electronic health record systems in hospitals/clinics. Portals typically involve allowing patients access to viewing their test results, visit summaries, and sometimes emailing with providers. Because most healthcare systems (with the exception of places like Kaiser) have only just recently created these portal websites as a part of the Meaningful Use certification process, there is a huge need for research studying their use and effectiveness. 

This post includes my thoughts about what we know and don’t know from the research to date, especially in light of two recent systematic reviews on 1) the effect of portals on outcomes  and 2) how patients with chronic disease and their providers view using an online portal.  (Apologies in advance that it is a very research-specific post!)

·         Whites, women, individuals younger than 65 years old, and those with higher education use portals the most.  This finding is consistent across virtually every study and every healthcare system – think it is very well-established that there are large socio-demographic differences in use.  I think the more important issues to highlight are: 1) across patient populations – including here at SFGH – the interest level is high for accessing information online and communicating with providers electronically, and 2) racial/ethnic differences in portal use persist even when adjusting for everyday Internet use. Therefore, we have to look beyond interest and access as the sole explanatory factors to be able to figure out how to engage diverse groups in signing up and logging in.

·         89% of the portal studies were non-experimental and none were randomized controlled trials (RCT). There is a need for more methodological rigor when studying online patient portals.  But it makes sense that no one published an RCT – the gold standard of effectiveness evidence – about portals in the past 5 years: these websites are being implemented as a part of standard care, and no system is willing to delay implementation to be able to design and implement an RCT when there are billions of financial incentives on the line.  However, there are many quasi-experimental designs that can inform our understanding about the link between portals and health and healthcare outcomes; we need to go beyond cross-sectional assessments. 

·         There are many ways in which we can improve the exposure and outcome measurement for these studies.  We should start to think of portal use as a set of various health behaviors rather than “any vs. no” use.  Some healthcare coordination tasks like making appointments may be very distinct from logging on to email a provider about medication side effects (and may require different skill sets to complete).  Similarly, since portals are linked to EHRs, let’s ensure that we choose clinical and/or utilization measures from the medical record that are conceptually linked to use of specific features of the portal.  For example, we have recently published on use of the online refill function in relation to statin adherence.

·         “A recurring theme in the literature is the inability of patients to understand medical terminology presented in the patient portal and not being knowledgeable about their own condition.”  This is a direct quote from the second review article, and it cannot be overstated enough in my opinion.  We are moving forward rapidly to expand implementation of online patient portals, but there is very little to no work to ensure that patients can actually use the information provided on these sites.  This is even more true in a setting like SFGH, in which many of our patients face language and literacy barriers.  We need to employ new design strategies to address usability upfront. 

As researchers, clinicians, or healthcare administrators, everyone in the healthcare setting will be shifting the standard of care to use portal websites to communicate with patients in the coming years.  Let’s all work together to design and implement studies that will help us be able to do this most effectively and with all of patients who are interested in accessing their information online.