Providing patients access to the electronic health record (EHR) through the internet web-based portals has been demonstrated to facilitate patient engagement, improve patient and health provider communication, increase patient satisfaction from care received, and result in better clinical outcomes (Mendu & Waikar, 2015; Nguyen, Bartlett, Rodriguez, & Tellier, 2016).
Several research studies have identified possible barriers to patient portal adoption including differences in internet access, computer literacy, and internet proficiency, often referred to as the digital divide (Jhamb et al., 2015; McCloud, Okechukwu, Sorensen, & Viswanath, 2016). The digital divide is one explanation for underserved groups being less likely to use portals, with strong correlations to utilization by race and household income (Jhamb et al., 2015; Kushinka, 2013; McCloud et al., 2016). With the increasing emphasis on Health Information Technology applications, already disadvantaged groups may be at further risk of poorer health outcomes if they lack access to HIT (Mendu & Waikar, 2015).
As smart phones proliferate and data access costs decline, it is important to conduct a current evaluation of patient access to and familiarity with the internet to use the portal, especially among patients from lower socioeconomic levels. The study was designed to quantify whether patients attending a community clinic had the necessary technological gateway, as well as internet experience to be able to use the patient portal following access activation. In addition to internet access, the study looked at educational level, household income, and self-perception of health as covariates with previous and future portal use.
Methods:
Setting: A family clinic providing medical care, family planning education, and psychological counseling. Among the overall clinic population, approximately 65% of patients report having government-supported insurance; 3% reported no insurance coverage. Forty-five percent of patients self-report as Latino and 25% Asian/Pacific Islander.
Instrument: We adapted the questionnaire designed by Czaja (Taha, Sharit, & Czaja, 2014) to document portal usability barriers to a one page, two-sided survey with English on one side and Spanish on the other. The questionnaire took 3-5 minutes to complete and consisted of demographics, internet use, past portal use, and future portal use.
Recruitment: From March 2017 to October 2017, a bilingual, bicultural research associate recruited potential participants from a convenience sample as they waited for a regularly scheduled clinic visit. Participants were eligible if they were receiving care from the clinic, were 18 or older, were English or Spanish readers, and were willing to complete the survey.
Results:
Participants Demographics: One hundred fifteen participants ranging in age from 18 to 84 (mean 42.1) completed the survey. Two participants (1.5%) completed the survey in Spanish. Thirty-five (30%) of the participants self-identified as Latino; 12 (10%) as Asian; and 20 (17%) as other. Almost 80% reported their health as good or better. Although 38% reported completing some college and 47% reported being college graduates, 60% reported a household income of less than $50,000.
Technology Use: Within the sample 87% had used the internet for more than a year and 60% reported using the internet for 5 or more hours a week. Portal Use: 47 (42%) had used the portal previously. There were significant differences in previous portal use by reported time spent on the internet during the week (Fisher’s Exact = 9.59; p = .02) and smart phone internet access (Fisher’s Exact = 6.15; p = .02). Using a computer for internet access was significantly associated with income level (Fisher’s Exact = 16.91; p<.001). Reported use varied from looking at test results (31%); secure communication with providers (20%); scheduling appointments (18%); and requesting prescription refills (9%). Ninety respondents (83%) would like to use the portal in the future, primarily to view test results (79%), secure communication with provider (59%); and prescription refills (59%). A logistic regression was used to determine the relationship of the significant predictors for future portal use of phone internet access, computer access, self-reported health status, sex, and age. With an overall chi-square of 22.90 (p=.001), a Naglekerke R-squared of 0.33; and a classification of 86%, computer internet access (OR 4.3 (95% CI: 1.15, 16.3)) was the only significant variable in the model, when controlling for the other independent variables.
Conclusions:
Among this highly educated but lower economic sample, there was a high level of access to the internet as well as weekly use of the internet. Within the sample, use and technological gateway modality were not only associated with past use but also the likelihood of future use. There are limitations to this study including convenience sample recruitment and the fact that the race/ethnicity profile of participation do not match the clinic’s demographics. Although there was no difference in past or predicted use of portal by income, the significance of using the computer, which is associated with higher income, to access the portal may reflect less restricted internet access without data allowances and the need to be able to interact with the relatively complex medical record on a screen that is large enough to see the information in a single view without scrolling.