Characteristics of older adults using patient web portals to view their DXA results

In 2009, the Centers for Medicare and Medicaid Services (CMS) established the Electronic Health Records Incentive Programs “to promote the adoption and meaningful use of interoperable health information technology (HIT) and qualified electronic health records (EHRs)” [1]. One HIT approach for achieving this goal is sharing information in the EHR with patients via web portals. The advantages of providing patients with access to their test results are numerous, including fostering self-efficacy and patient activation as well as preventing test results from falling-through-the cracks, which is a common medical error and a threat to patient safety [2]. For this reason, the Joint Commission established goals for communicating test results [3]. Patient portals, like MyChart from Epic, is one method patients can access their test results. Patient web portals have been shown to improve patient satisfaction and health care quality [4‐10].

A commonly performed test among older adults is dual energy x-ray absorptiometry (DXA), which is the standard screening tool for identifying patients at increased risk for hip or other fragility fractures. But even when a diagnosis of osteoporosis is made from the DXA, patients do not always receive appropriate pharmacotherapy consistent with treatment guidelines [11]. Indeed, one study found that two-thirds of patients with DXA results indicating osteoporosis did not receive appropriate pharmacotherapy. And using chart review methods, another study revealed among 23% of abnormal DXA results there was no evidence that the patient’s clinician had reviewed their DXA results or communicated them to their patients [11]. One way to increase patient access to their DXA results would be to use web portals to communicate test results directly to patients. To our knowledge, however, there have not been any studies of patients’ use of web portals for learning about their DXA results. Sharing DXA results with patients via web portals may also activate them to follow-up with their health care provider, especially if their results are abnormal.

Web portals are relatively new in medicine. A recent systematic review found low use rates for patient web portals across a number of studies [12]. The main reasons given included concerns about confidentiality, poor awareness of web portals, and inadequate user friendliness. Overall, patients who did use web portals as part of the health care were typically younger, and more likely to be White, affluent, women with chronic disease [12].

Our objective was to examine patient and clinical factors associated with the use of web portals to review their DXA results. We did this using data from the Patient Activation after DXA Result Notification clinical trial (PAADRN- https://​clinicaltrials.​gov/​ct2/​show/​NCT01507662), Identifying the characteristics of web portal users, specifically for DXA results, may provide the information needed to inform practitioners about the appropriate ways to communicate DXA results, as well as provide insight into how best to craft interventions to increase web portal use.

We used data from a large multi-site, pragmatic randomized controlled trial to examine which patients learned of their DXA results using available patient web portals. Our analysis revealed that most participants (80.9%) never used the patient web portals to view their DXA results. We also found that portal use was more common among patients who were at UI, younger than 75, more educated, had higher health literacy, self-reported osteopenia, and had a provider of the same gender.

These findings are comparable to reports from other studies that have found web portal use rates ranging from 16 to 35% [22‐28]. Direct comparisons are difficult, however, because of the varying definitions of web portal use including whether patients were only registered [22, 26, 28], or if they logged in, or if they activated their account [23‐25, 27]. Our study was restricted to patients who reported knowing their DXA results and said that they viewed their DXA results using the web portal. We were not, however, able to know if those who did not use the web portal to view their DXA results used the portal for other purposes.

Low rates of web portal use may be due to patients preferring other forms of communication with their providers for getting their test results. While web portals are strongly endorsed [1], patients may not like them [29]. A study of patient preferences for DXA result notification found that 18% of the patients considered it unacceptable to provide either normal or abnormal DXA results over a secure web portal [29]. Perhaps when patients become more familiar with web portals, their acceptance of web portal notification of test results will increase. Learning more about patient’s perceptions of web portals may also lead to increased use, especially among older adults.

We found that after adjusting for other factors, patients > 75 years old had 67% lower odds of using the web portal to view their DXA results. This is consistent with most prior studies of web portal use [22], although Krist et al. reported an anomalous finding that patients 60–69 years old were more likely to use the web portal than patients 18–59 or 70 years old or older [24]. It has been suggested that older patients actually may be more engaged to use a web portal if they have more chronic diseases, office visits, and diagnostic tests to be informed about [24, 30, 31]. For example, Ancker et al. did find that those with chronic conditions were more likely to register for a patient web portal [22]. While we did not find a significant difference in web portal use based on the number of patient comorbidities, we did find that those who self-reported osteopenia had 45% higher odds of viewing their DXA via their web portal results than those with normal bone density. Patients with a history of osteopenia may be more anxious and thus more motivated to get their DXA results.

Unlike other studies [22‐24, 26‐28], we did not find that race or sex differed between web portal users and non-users. This may be due to the low number of non-Whites, especially at UI, and to having men included in our sample given that osteoporosis is general associated with White females. One of our most interesting findings was that patients with the same gender as their provider had 41% greater odds of using the web portal than those with a provider of a different gender. To our knowledge, only one other study has examined clinician characteristics related to patient web portal use, and they also found that patients with female clinicians were 37% more likely to use a web portal [24]. Others have shown that women are more likely to be earlier adopters of web portals than men [22‐24], and clinician adoption of web portals may stimulate patient use [24].

Consistent with other studies [12, 26, 32], we found that more educated patients had between 56 and 81% greater odds and those with higher health literacy had 50% greater odds of using the web portal to view their DXA results than those with lower education and health literacy. We expected this finding because activating, accessing, and navigating a web portal requires patients to read health-related terms. If web portals are to be emphasized and encouraged going forward, developers should keep in mind health literacy when designing web portal interfaces and how test results are communicated to patients.

Finally, there are factors that limit the generalizability of our findings. First, we included 50–64 year olds even though DXA is generally performed on older adults. In a sensitivity analysis restricted to patients > 65 years old (data not shown), however, the associations reported here were comparable although statistical significance may not have been achieved given the reduced statistical power. Second, we relied on patient-reports of how they learned about their DXA results. Due to IRB constraints and time limitations of the two sites’ information technology staff, we were unable to confirm the validity of those patient reports. Third, those who received the intervention DXA result letter may not have accessed their information via a portal because they already received their results. However, we waited four weeks after the participant had their DXA before mailing the result letter to see if they would to allow time for them to attempt to get their results in another manner. Lastly, because we could not determine the number of participants who had activated web portal accounts, the rate of those who did access their DXA results (13.9%) may have been higher if we looked at only those with activated accounts. However, those who did not have an activated account are still considered non-users of web portals and as such did not view their DXA results on the web portal.

Patient web portals offer an efficient and quick vehicle through which clinicians can communicate with patients, and for patients to be more actively involved in their health care. Determining which patients use web portals to view their DXA results is an important step in encouraging web portal use for health communication. We found that older patients and those with less education or lower health literacy or lower levels of patient activation were less likely to use the web portal. Future research should examine the barriers and facilitators for patients accessing web portals and determine their preferences for receiving their DXA results.