Electronic Health Record Patient Portals in CKD and Hypertension Management: Meaningfully Used?

Providing patients access to their personal health records by means of web-based portals to facilitate patient engagement has been well established for various chronic conditions, including diabetes and congestive heart failure (1–3). The most successful of these interventions have been shown to improve patient-provider communication, patient satisfaction, and clinical outcomes (4). In addition, various forms of technology, including the use of mobile devices, personal computer–based platforms, and tablet-based platforms, have been used to improve outreach among patients with complex medical issues (5,6). Although few studies have investigated the use of technology to improve engagement among patients with kidney disease, there is evidence that these patients are interested in self-management through web-based platforms (7,8). Hypertension management—because of its high prevalence (9), importance in CKD treatment (10), and need for patient engagement—may be particularly well suited for technologic innovation.

In their article “Disparities in Electronic Health Record Patient Portal Use in Nephrology Clinics,” Jhamb et al. (11) examined the use of an electronic health record (EHR) portal among patients cared for by four university–affiliated nephrology practices, including 24 nephrologists. Jhamb et al. (11) report results on two related but conceptually distinct domains: (1) descriptive analyses of secular trends and socioeconomic determinants of use (i.e., was the portal used less by those who were poorer, older, or from underserved minorities?) and (2) analyses of the effect of the portal on BP control (i.e., did the portal make a difference on probably the most important modifiable risk factor in CKD management?). Jhamb et al. (11) succeed on the first domain by showing stark disparities in portal adoption, but ultimately fail to make the case that the portal influenced clinical outcomes. The report is nevertheless important and timely and highlights action items for the optimal use of technology in CKD care. A brief review of the portal and study design will serve to show the study’s strengths and limitations.

Over a 3-year period, several months after portal launch, Jhamb et al. (11) studied whether and how the portal was used by 2803 patients. Patients were invited to sign up for the portal by staff during appointment check in, and 5th to 6th grade–level materials were displayed. Portal enrollment involved patients obtaining an access code, going online, and following instructions to set up a secure account. The portal itself was predominantly a gateway for patients to convey requests for appointments and medication refills and review laboratory results. An important feature of the portal included emailed alerts about new information, results, or appointments. Unfortunately, fewer than one half (39%) accessed the portal; >80% used it for scheduling appointments or reviewing laboratory test results. Medications were reviewed by 77%, and 65% requested refills. Fewer than one third used the portal to communicate with their nephrologist regarding medical advice. Jhamb et al. found that the use of the portal increased over time and was more commonly accessed by generally healthier patients and those at lower risk of CKD progression. Portal users were younger, were one half as likely to be black, were more likely to be married, were more likely to have private health insurance, and had higher median incomes.

The strengths of this study include its novel assessment of EHR portal use in patients cared for by nephrologists, with a focus on demographic factors contributing to decreased adoption. As Jhamb et al. point out, it is not surprising that disadvantaged patient populations are at risk for portal adoption failure and potentially worse clinical outcomes given the established disparities in both access to technology (12) and delivery of kidney disease care among these patients (13). Shedding light on how implementation of technology–based care innovations can introduce further inequities in care delivery for a condition where stark disparities already exist is an important finding of the study by Jhamb et al. The study shows that portals could perversely widen existing disparities in care by advantaging those who are already at an advantage, while not helping the disadvantaged (11). These findings illustrate the importance of incorporating strategies inclusive of vulnerable patient populations when implementing care delivery interventions. For example, surveying patients beforehand about access to portal technology, incorporating a smartphone application, allowing patients to sign up for access, and showing portal usability within the clinic may lead to a better understanding of portal adoption barriers and potentially improve usage among disadvantaged patients. In addition, patient-centered outcomes that are of value to patients should be an important measure for future studies of technologic innovations in health care. In this study, it is possible that nonportal users placed less value on what the portal offered. By addressing factors that relate to quality of life and elements of care most important to patients, future investigators can potentially mitigate disparities in technology–based care delivery.

There are a number of limitations of this retrospective single–center cohort study. First, there is limited generalizability given that the four nephrology practices were located in western Pennsylvania and affiliated with a single university. Second, there were key baseline differences in health status between nonportal users and portal users, who were healthier with lower BP, higher eGFR, and lower rates of coronary artery disease and diabetes. The mixed effect logistic regression model adjusted for some but not all of the baseline imbalances. As with any observational study, there remains the strong possibility of residual confounding from difficult to quantify variables such as health literacy, educational status, and access to technology. Third, the use of BP control as a dichotomous variable as opposed to examining change in BP over time is a limitation. Supplemental figure 1 in Jhamb et al. (11) illustrates nearly parallel lines of BP control between portal users and nonusers. This raises the question of whether the change in BP control was any different (or even lower among portal users) given that baseline BP control seemed to be better among portal users. In addition, the study used clinic BPs as opposed to home BP recordings to capture and assess BP control. Greenberg et al. (14) reported that incorporation of BP recordings from free text notes and home measurements significantly affected the measured rate of control compared with EHR BP fields alone.

Finally, the EHR portal itself was relatively limited in its capabilities. There were no educational materials related to kidney disease health or BP control, and there were no interactive features allowing patients to enter health information and monitor trends over time. Future implementation of a more comprehensive, multifaceted portal–based intervention could facilitate patient education, serve as an interactive tool to track BPs over time, and enable bidirectional communication about medication changes and adherence. This could more plausibly achieve the goal of improving BP control for patients with kidney disease.

Numerous patient– and provider–related factors contributing to low rates of BP control have been elucidated (15). Patient factors include race, age, obesity, limited access to health care, knowledge deficits, poor communication with physicians, cost of care, and lack of compliance with treatment or diet. Provider factors include lack of knowledge or agreement about guidelines and therapeutic inertia (failure to increase therapy in the setting of poor BP control). Clearly, some of the variables are nonmodifiable, but some, such as patient knowledge deficits, poor communication between patients and physicians, and physician understanding of guidelines, can and should be addressed at a systems level. An EHR-based portal can undoubtedly address some of these barriers to hypertension care delivery, such as patient education and communication. Quality improvement studies to date that have succeeded in achieving an appreciable clinically significant reduction in BP have included patient education, frequent and close monitoring of real–time BP recordings, and increased communication with patients (often through a medical assistant, nurse practitioner, or pharmacist) (16–18). These studies suggest that a technology–based patient–centered portal incorporating various elements that address traditional barriers to BP control has the potential to improve clinical outcomes. Fortunately, the National Institutes of Health is supporting research examining the use of the EHR in guiding hypertension and CKD management, which could spur further innovation in this field (19).

Jhamb et al. (11) have taken an important first step forward in investigating the use of an EHR-based portal to improve BP control. They have illustrated the importance of accounting for vulnerable patient populations in designing technology-based interventions for patients with kidney disease. It is imperative that nephrologists explore innovative solutions focused on barriers to care to improve the health and wellbeing of patients with hypertension and kidney disease, particularly in an age of evolving technologies that can be harnessed to improve health care.

Disclosures

None.