Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 8.
Published in final edited form as: Res Gerontol Nurs. 2020 Sep 23;14(1):33–42. doi: 10.3928/19404921-20200918-03

Predictors of patient portal use among community-dwelling older adults

Maan Isabella Cajita 1, Marci Lee Nilsen 2, Taya Irizarry 3, Judith A Callan 4, Scott R Beach 5, Ellen Swartwout 6, Laurel Person Mecca 7, Richard Schulz 8, Annette DeVito Dabbs 9
PMCID: PMC8992382  NIHMSID: NIHMS1784784  PMID: 32966584

Abstract

Older adults lag behind younger counterparts in the use of patient portals, which may limit their ability to engage in health care. A better understanding of the factors associated with portal use among older adults is needed. We examined the proportion of 100 community-dwelling older adults who reported using a portal, the associations between socio-behavioral factors and portal use and modeled predictors of portal use. Of the 52% who reported using a portal, 28% used the portal on their own, and 24% received assistance from others or had others access the portal on their behalf. After controlling for confounders, only marital status was significantly associated with any portal use. Marital status and patient activation were significantly associated with independent portal use. Further exploration is warranted to identify additional factors and the possible mechanisms underlying portal use by older adults.

Keywords: patient portal, electronic health record, technology, older adults

Introduction

A well-informed patient is an empowered patient, and patient portals provide an excellent way for patients to keep up to date with their health information and manage aspects of their health. Patient portals are two-way, Internet-based channels for communication between patients and health providers, tethered to the provider-maintained electronic health record (EHR) (Irizarry, DeVito Dabbs, & Curran, 2015). Since 2014, United States (US) health care providers have been required to provide patients not only with access to their electronic health information but also a secure means of communicating with providers (Nahm et al., 2018), and patient portals have emerged as the most common vehicle for health providers to demonstrate compliance with meaningful use requirements. Other countries—including Denmark, Finland, United Kingdom, and Australia—also provide patients access to their EHRs via portal (Rigby et al., 2015).

Unfortunately, the use of patient portals has remained low (Rigby et al., 2015). Although patient portal availability increased 10% (from 42% to 52%) between 2014 and 2017, the rate of using the portals among patients rose by only 1.2% (from 26.8% to 28%) (Nahm, Sagherian, & Zhu, 2016; Patel & Johnson, 2018). Furthermore, of the 28% of all patients who use patient portals as a tool for health care engagement, only a fraction are over 65 years of age (Patel & Johnson, 2018).

Older adults, who utilize the greatest proportion of health care resources, often face difficulties using patient portals—particularly those older adults who have lower numeracy skills and less experience with technology (Taha, Sharit, & Czaja, 2014; Zarcadoolas, Vaughon, Czaja, Levy, & Rockoff, 2013). Other factors that prevent older adults from adopting patient portals include lack of access to technology and the Internet, lack of computer and/or Internet skills, visual and cognitive impairments, decreased function and dexterity of the upper extremities, and concerns over the security and privacy of their health information (Sakaguchi-Tang, Bosold, Choi, & Turner, 2017). These factors have been implicated in the emergence of the grey digital divide (i.e., the gap between those who have ready access to technology and the skills to make use of those technologies and the older adults who do not) (Anderson & Perrin, 2017), which places older adults at a disadvantage in health care engagement that is facilitated electronically.

On the other hand, factors that have been shown to facilitate patient portal use among older adults include doctor’s or family member’s recommendation and receiving technical assistance (Sakaguchi-Tang et al., 2017). In addition, person-level factors such as age, ethnicity, education level, health status, and health literacy, or the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions (Health Literacy: A Prescription to End Confusion, 2004), have been shown to influence the patient’s interest and ability to use patient portals (Irizarry et al., 2015; Powell, 2017). Other factors associated with patient portal use are experience in using computer technology (Latulipe et al., 2015) and patient activation, or the knowledge, skills, and confidence an individual has in managing their health (Hibbard, Stockard, Mahoney, & Tusler, 2004; Smith, Pandit, Rush, Wolf, & Simon, 2015).

Despite the increasing number of studies that investigate how patients adopt electronic patient portals, little is known about the socio-behavioral factors that promote patient portal uptake among older adults. A better understanding of these factors would help inform the design and implementation of patient portals. Therefore, the objectives of this study were to: (1) identify the proportion of community-dwelling older adults who reported using a patient portal, (2) examine the associations between socio-behavioral factors (e.g., socio-demographics, health status, patient activation, health literacy, and technology experience) and patient portal use, and (3) model predictors of portal use among community-dwelling older adults.

Methods

Study Design and Sample

This study featured a cross-sectional, correlational design. Convenience sampling was used to recruit a sample of community-dwelling, English-speaking, older adults (≥65 years of age) from the University of Pittsburgh, University Center for Social and Urban Research (UCSUR) registry. The demographically diverse registry contained approximately 9,000 regional residents of the Pittsburgh Metropolitan Statistical Area who were willing to be contacted for participation in research studies. The participants included in the registry had participated in a variety of population-based surveys conducted by UCSUR and were found to be largely representative of the regional population and reflected adequate variation in socio-behavioral characteristics of interest (e.g., education level, health literacy). To be eligible for participation in the study, the participants recruited from the UCSUR registry were at least 65 years of age or older and had to be community-dwelling. The study was approved by the University of Pittsburgh, Human Research Protection Office, Institutional Review Board.

Procedures

Registry personnel contacted a total of 161 older adults about participating in the study—of whom 52 were unreachable, nine declined to participate, and the remaining 100 individuals agreed to participate (response rate of 62%). Each potential participant was then mailed a copy of the study’s consent form. Prior to administering the 45-minute telephone survey, the interviewer obtained the participant’s verbal consent in accordance with IRB-approved guidelines. Each participant received $10 for participating.

Measures

The survey measures were selected from the core battery for demographics, health, and disability from the Quality of Life Technology Engineering Research Center (NSF—0540865) and the Center for Research and Education in Aging and Technology Enhancement (CREATE) (NIH—5P01AG017211-21), which included the Institute of Medicine’s proposed socio-behavioral factors for inclusion in electronic health records (Czaja et al., 2006).

Portal use

Use of a patient portal was determined by the participant’s responses to the following questions: (1) Have you ever used a patient portal? (Yes / No); (2) Does someone who helps you with your health, help you to use the portal? (Yes—all the time / Yes—some of the time / No); (3) Does someone who helps you with your health, access the portal on your behalf? (Yes—all the time / Yes—some of the time / No). Participants who answered “Yes” to any of these questions were categorized as patient portal users (with or without assistance).

Socio-demographic Factors

Variables included age, sex, race, marital status, and educational attainment. Age was treated as a continuous variable. Race was dichotomized as either white or other. Similarly, marital status was dichotomized as (i.e., either single / divorced / widowed or married / living with a significant other). Lastly, educational attainment was dichotomized as either having less than a college degree or having a college degree or greater.

Health Status

General health status was measured using the single-item General Self-Rated Health (GSRH) measure (DeSalvo et al., 2006). Participants responded to “In general, how would you say your health is?” according to a 5-point Likert scale (Poor = 0, Fair = 1, Good = 2, Very good = 3, or Excellent = 4). Per convention, responses were then dichotomized to Poor/Fair and Good/Very good/Excellent (Rosenzveig, Kuspinar, Daskalopoulou, & Mayo, 2014). The GSRH has been validated among veterans and has shown robust reproducibility, reliability, and validity (DeSalvo et al., 2006).

Patient Activation

The 13-item version of the Patient Activation Measure (PAM-13) was used to measure patient activation (Hibbard et al., 2004). Possible PAM-13 scores range from 0 to 100 and are categorized into four levels: Level one (score ≤ 47—respondents believe in taking an active role but are unprepared; Level two (score 47.1–55.1—respondents have some knowledge but still struggle to manage their health; Level three (55.2–67—respondents begin to take action but do not have the skills to sustain their behavior; Level four (score ≥ 67.1—respondents can sustain self-management behaviors, even while under stress (Hibbard, Mahoney, Stockard, & Tusler, 2005). The PAM-13 has demonstrated good internal consistency (Cronbach’s alpha = 0.9) and construct validity in studies of older adults (Skolasky et al., 2011).

Health Literacy

A single-item of the Brief Health Literacy Screen (BHLS) (How confident are you filling out medical forms by yourself?) was used to measure the adequacy of health literacy (Chew, Bradley, & Boyko, 2004). Participants responded using a 5-point Likert scale (Extremely = 0, Quite a bit = 1, Somewhat = 2, A little bit = 3, and Not at all = 4). Using the suggested threshold (Chew et al., 2004), participants who scored two or higher were deemed to exhibit inadequate health literacy. The ability of this single BHLS question to detect inadequate health literacy compared to two standard measures of inadequate health literacy (S-TOFHLA and REALM) has been established (among a group of 1,259 veterans aged 50 years or older), with the area beneath the receiver operating characteristic curve ranging from 0.7 to 0.8 (Chew et al., 2008).

Experience with Technology

The participants were asked whether they used a cellphone/smartphone, computer, or home device (e.g. security systems, remote appliance setting) for health-related activities. From this question, a dichotomous variable labelled “health-related technology use” was created. Next, the participants were asked whether they have searched online for health information (e.g., medication information, availability of health services, information about health professionals, and/or information about health care facilities). From the second question, a dichotomous variable labelled “health-related online use” was created.

Statistical Analyses

Data were analyzed using Stata/SE 15 (StataCorp LP, College Station, Texas, USA). Descriptive statistics were calculated for all variables (mean, standard deviation, and proportions). Only two variables (race and education level) exhibited missing data (one missing data point each). As such, Little’s test was performed to check the pattern of the missing data, and a highly nonsignificant p-value (p = 0.4) indicated that the data were missing at random. Participant characteristics between those who used the patient portal and those who did not were compared with a Pearson’s chi-squared test (χ2) (for categorical variables) and a two-sample T-test (for continuous variables). Associations between socio-behavioral factors were estimated using a Kendall’s rank correlation test with Bonferroni correction. Additionally, the multicollinearity of factors was assessed with a variance inflation factor (VIF). The resultant VIFs ranged from 1.2 to 1.3. The mean VIF was 1.2, which indicated a weak correlation among factors. To model predictors of portal use (with or without assistance), first, bivariate logistic regression analyses were conducted to identify potential correlates (cut-off threshold p ≤ 0.05). Then, a multivariate logistic model was used to identify significant predictors of patient portal use (p ≤ 0.05). As a final step, the fit of the model was tested with a Pearson’s Goodness-of-fit test.

Results

Sample Characteristics

Socio-behavioral characteristics of participants in total and by portal use are shown in Table 1. A total of 100 older adults participated in the study. The mean age of the participants was 74.7 ± 1.2 years, and 58 per cent were female. The majority (78%) identified themselves as White. Thirty-nine per cent possessed at least a college degree and 48% were married or living with a significant other. For reference, among the US population aged 65 years and older, 86% identify as White, 34% have a college degree or higher, and 57% are married (U.S. Census Bureau 2018). The majority (69%) of the participants rated their general health as good to excellent. Indeed, almost half (49%) of the participants reported the ability to sustain self-management behaviors, even when under stress (based on meeting the threshold of level four for high patient activation on the PAM). An even greater number of participants (75%) reported possessing adequate health literacy. Of central importance to this study, 76% of participants report using technology for health-related activities and 56% reported going online to search for health-related information (see Table 1).

Table 1.

Socio-behavioral characteristics of participants in total and by portal use

Total
N = 100		 Uses portal
n = 52		 Does not use portal
n = 48		 X 2 p-value
SOCIO-DEMOGRAPHICS
Age, mean (SD) 74.7 (1.2) 75.2 (0.99) 74.2 (1.1) 4.6 # < 0.001
Sex, N (%) 0.6 0.5
 Male 42 (42%) 20 (38%) 22(46%)
 Female 58 (58%) 32 (62%) 26 (54%)
Race, N (%) 0.2 0.6
 White 78 (79%) 40 (77%) 38 (81%)
 Other 21 (21%) 12 (23%) 9 (19%)
Marital status, N (%) 8 0.005
 Single/divorced/widowed 52 (52%) 20 (38%) 32 (67%)
 Married/living with significant other 48 (48%) 32 (62%) 16 (33%)
Education, N (%) 0.1 0.7
 Less than a college degree 60 (61%) 30 (59%) 30 (63%)
 College degree or greater 39 (39%) 21 (41%) 18 (37%)
HEALTH STATUS
General health status, N (%) 0.2 0.7
 Poor to fair 31 (31%) 17 (33%) 14 (29%)
 Good to excellent 69 (69%) 35 (67%) 34 (71%)
PATIENT ACTIVATION
Patient activation, N (%) 1 0.3
 Levels 1–3 (score ≤ 67) 51 (51%) 24 (46%) 27 (56%)
 Level 4 (score > 67) 49 (49%) 28 (54%) 21 (44%)
HEALTH LITERACY
Health literacy, N (%) 1.9 0.2
 Inadequate 25 (25%) 10 (19%) 15 (31%)
 Adequate 75 (75%) 42 (81%) 33 (69%)
EXPERIENCE WITH TECHNOLOGY
Health-related tech use, N (%) 4.4 0.04
 No 24 (24%) 8 (15%) 16 (33%)
 Yes 76 (76%) 44 (85%) 32 (67%)
Health-related online use, N (%) 2.5 0.1
 No 44 (44%) 19 (37%) 25 (52%)
 Yes 56 (56%) 33 (63%) 23 (48%)
Open in a new tab

Note

#

= T-test.

Patient Portal Use

Fifty-two participants reported that they used a patient portal (with or without assistance). Of these, 28 used the portal independently; six received assistance from others; five had others access the portal on their behalf;13 received assistance from others and/or had others access the portal on their behalf. Among the participants who received assistance in using the patient portal, three reported receiving assistance all the time and 16 only some of the time. Among those who had someone else access the portal on their behalf, six reported having someone else access their portal all the time and 12 only some of the time.

Associations between Socio-behavioral Factors

We observed evidence of moderate positive associations between health literacy and patient activation (tau = 0.4, p < 0.001), marital status and health-related technology use (tau = 0.4, p = 0.03), and health-related online use and health-related technology use (tau = 0.4, p = 0.02). The associations between the remaining socio-behavioral factors (see Table 2) exhibited no statistical significance (Khamis, 2008).

Table 2.

Associations between socio-behavioral factors

Age Sex Race Marital status Education Health status Patient activation Health literacy Health-related tech use Health-related online use
Age 1
Sex −.004 1
Race −.06 .2 1
Marital status −.3 .2 .3 1
Education −.07 .2 .1 .05 1
General health status .1 .2 .2 .05 .1 1
Patient activation −.1 −.1 −.1 .02 −.1 .2 1
Health literacy −.1 .01 .05 .1 .03 .2 .43 1
Health-related tech use −.2 −.1 −.05 .35 .3 .2 .2 .2 1
Health-related online use −.2 −.06 .01 .2 .2 .04 .04 .1 .36 1
Open in a new tab

Note: Kendall’s rank correlation test with Bonferroni correction

Bivariate Associations between Socio-behavioral Factors and Portal Use

Participants, who were married or living with a significant other, exhibited higher odds (odds ratio [OR] = 3.2, p = 0.005) of using a patient portal (with or without assistance) compared to participants who were single, divorced, or widowed. Participants who used technology for health-related activities exhibited higher odds (OR = 2.8, p = 0.04) of using a patient portal (with or without assistance) compared to those who did not.

Subsequently, marital status (OR= 6.5, p < 0.001) and health-related technology use (OR = 12.7, p = 0.016) were associated with independent portal use. Similarly, participants who reported high patient activation (PAM Level four) had higher odds of using a patient portal by themselves compared to participants who reported low activation (OR = 3.7, P = 0.007). Participants who had adequate health literacy had higher odds of using a patient portal independently compared to those who had inadequate health literacy (OR = 3.7, P = 0.05). Lastly, participants who searched online for health-related information had higher odds of using a patient portal on their own (OR = 4.1, P = 0.006). (Table 3)

Table 3.

Associations between socio-behavioral factors and patient portal use

Any portal use Independent portal use
Bivariate Multivariate Bivariate Multivariate
Odds ratio p-value Odds ratio p-value Odds ratio p-value Odds ratio p-value
Age 1 0.5 -- -- 1 0.3 -- --
Sex 0.7 0.5 -- -- 0.7 0.4 -- --
Race 0.8 0.6 -- -- 1.9 0.3 -- --
Marital status 3.2 0.005 2.7 0.03 6.5 <0.001 5.6 0.003
Education 1.2 0.7 -- -- 1.8 0.2 -- --
General health status 0.9 0.7 -- -- 1.9 0.2 -- --
Patient activation 1.5 0.3 -- -- 3.7 0.007 3.6 0.03
Health literacy 1.9 0.2 -- -- 3.7 0.05 1.5 0.6
Health-related tech use 2.8 0.04 1.9 0.2 12.7 0.02 3.5 0.3
Health-related online use 1.9 0.1 -- -- 4.1 0.006 2.7 0.1
Open in a new tab

Note: Logistic regression

Multivariate Model of Portal Use

With p ≤ 0.05 as the cut-off, our final multivariate model comprised marital status and health-related technology use. Participants who used technology for health-related activities exhibited greater odds of using a patient portal (with or without assistance); however, this result was not statistically significant (OR = 1.9, p = 0.2). Only marital status remained statistically significant in the final multivariate model, and participants who were married or living with a significant other had greater odds (OR = 2.7, p = 0.03) of using a patient portal (with or without assistance). A Pearson’s goodness-of-fit test indicated the final multivariate model possessed a good fit (p = 0.9).

The final multivariate model for independent portal use comprised of marital status, health literacy, patient activation, health-related technology use, and health-related online use. Only marital status (OR = 5.6, p = 0.003) and patient activation (OR = 3.6, p = 0.031) were significantly associated with independent portal use in the final model. A Pearson’s goodness-of-fit test indicated the final multivariate model for independent portal use possessed a good fit (p = 0.5). (Table 3)

Discussion

Patient Portal Use

Among our sample of 100 community-dwelling older adults, 52% reported using a patient portal. This is higher than the proportion of the general public in the Health Information National Trends Survey (HINTS) who have reported accessing a patient portal (28%)(Patel & Johnson, 2018). The different sampling strategies employed in our study and in the HINTS could be a reason for the discrepancy in the prevalence of portal use. The HINTS used random sampling whereas convenience sampling was used in our study, which could have introduced selection bias, wherein older adults who used patient portals might be more likely to participate in a study about patient portals. Similar to our study, Nahm et al. (2016) reported that 60.6% of older adults, who they recruited from a senior’s online group (SeniorNet), used patient portals.

Associations between Socio-behavioral Factors

Similar to the authors of a prior study (Smith, Curtis, Wardle, von Wagner, & Wolf, 2013), we found a moderate positive association between health literacy and patient activation. However, contrary to our findings, Couture, Chouinard, Fortin, and Hudon (2018), found no relationship between health literacy and patient activation in a similar sample of adults with a mean age = 60 ± 13 years and at least one chronic disease. These conflicting findings may be due in part to the variety of health literacy measures, used among the studies (e.g. BHLS, Test of Functional Health Literacy in Adults [TOFHLA], and Newest Vital Sign [NVS]). Nevertheless, conceptually, health literacy and patient activation overlap to a certain degree; therefore, the moderate statistical association is not surprising (Hibbard, 2017).

The association between marital status and health-related technology use, albeit moderate, also was statistically significant, and the role of social influence could potentially explain this association. According to the Model of Technology in Households (MATH) (Brown, Venkatesh, & Bala, 2006), adoption of technology is influenced by the members of a given individual’s social network. For older individuals, because their households are typically comprised of their spouse or partner, their decision to use technology will be mainly influenced by that partner. In addition, we observed a moderate association between health-related technology use and health-related online use. Considering that most of the technologies included in our technology experience questionnaire were information and communication technologies (e.g., cellphones and computers), these technologies were associated, which should not be surprising because these technologies are commonly used to access the Internet (Anderson & Perrin, 2017).

Associations between Socio-behavioral Factors and Portal Use

At the bivariate level, only two of the socio-behavioral factors were significantly associated with any patient portal use, namely marital status and technology experience. However, marital status, health literacy, patient activation, and technology experience were associated with independent portal use. Participants who used technology for health-related activities were more likely to use a patient portal with or without assistance. Having technology experience has been found to be an essential factor in patient portal adoption among older adults (Latulipe et al., 2015). Latulipe et al. (2015) reported that older adults, who lacked experience with computers during their working years, demonstrated a lack of interest and confidence in using a patient portal. Moreover, these older adults preferred having an in-person interaction with their health care provider and were concerned that patient portals would eventually replace in-person visits (Latulipe et al., 2015).

Similarly, marital status was associated with patient portal use. In this study, participants who were married or living with a significant other had higher odds of using a patient portal compared to participants who were single, divorced, or widowed. This finding is in line with that of a previous study which found that older adults who were married were more likely to utilize the patient portal compared to those who were not married (Arcury et al., 2017). Specifically, they reported that marital status remained significantly associated with patient portal utilization even after adjusting for potential confounders such as race, education, comorbidity, insurance status, Internet use frequency, and geographic location. In a retrospective study of hospitalized cancer patients, marital status was also significantly associated with patient portal adoption (Aljabri et al., 2018). Married patients were 60 per cent more likely to use a patient portal compared to patients who were divorced, single, or widowed (Aljabri et al., 2018). Marital status could be considered a proxy for social influence. Similarly, marital status could also act as a proxy for social support. Social influence and social support could help explain the relationship between marital status and patient portal use. Social influence has been known to predict technology adoption (Venkatesh, Thong, & Xu, 2016). For an older adult, one’s household social network may mainly consist of one’s spouse or live-in partner (Brown et al., 2006).

Older adults are more likely to require assistance from others when learning how to use new technology (Anderson & Perrin, 2017). Hence, having a spouse/partner, especially one who is familiar with how to navigate a patient portal, could potentially facilitate patient portal use. On the other hand, it has also been reported that married individuals who use patient portals do so on behalf of their spouses and not just for their personal use (Powell & Myers, 2018). In their qualitative study, Powell and Myers (2018) reported that several patients mentioned accessing the patient portal as proxies for their spouses. These patients shared that they have taken the responsibility of keeping up with their own and their spouses’ health information (Powell & Myers, 2018). In this case, instead of being the recipient of their spouses’ support, the participants were the ones providing their spouses with support. Further research is needed to explore the mechanisms underlying the possible role of social support and social influence as it relates to patient portal use. Findings may inform the design of training and ongoing support for future older patient portal users.

Participants who reported higher levels of patient activation, meaning they were more engaged in their health care, were more likely to use patient portals on their own. This finding is similar to that of a national survey of U.S. adults, which reported that the respondents who had high patient activation were more likely to access their medical records online (Smith et al., 2015). Given this conceptual definition of patient activation, it is reasonable to expect that older adults who have high levels of patient activation may be more likely to access a patient portal, a tool intended to assist patients in managing their health. It is interesting to note that patient portals when designed properly, can increase patient activation, suggesting a two-way relationship between patient activation and patient portal use (Solomon, Wagner, & Goes, 2012). Electronic portals enable patients to access their health information, communicate with their health care providers, and perform other health-related tasks, such as request prescription refills and schedule appointments (Patel & Johnson, 2018). In a way, portals provide patients with another avenue to engage in their health care, which could increase patient activation. Like in a recent study of hospitalized adults, wherein the introduction of a patient portal intervention led to an increase in patient activation (Schnock et al., 2019).

Similarly, participants who had adequate health literacy were more likely to use a patient portal on their own compared to their counterparts who had inadequate health literacy. This finding is in line with that of Smith et al. (2015), who explored patient portal use among older adults from the Health Literacy and Cognitive Function among Older Adults (LitCog) cohort. Levy, Janke, and Langa (2015) explored the relationship between health literacy and using the Internet to obtain health information and found that older adults with low health literacy were less likely to search for health information online compared to older adults with adequate health literacy. Health literacy is an important skill to make full use of a patient portal. Hence, it is not surprising that those with adequate health literacy were more likely to use a patient portal by themselves.

Unlike previous research, in this study there was not enough evidence to support the association between educational attainment and patient portal use. In a previous study, participants with greater than a high school education were shown to be more likely to access a patient portal compared to those with less education (Arcury et al., 2017). The impact of educational attainment on patient portal use might not be as significant among older adults due to a cohort effect. It was not until the early 1990s that the internet was made public (Conseil Européen pour la Recherche Nucléaire, n.d.), by which time today’s older adults had completed their formal schooling. Latulipe et al. (2015) noted that older adults’ lack of interest in using patient portals could be linked to the absence of computing technology during their formative and working years.

Limitations

This study might have been underpowered due to its relatively small sample size even though we included at least ten participants per covariate. Additionally, the participants were recruited from one geographic area, and minority ethnic groups were under-represented, which could limit the generalizability of our findings. As previously noted, the use of convenience sampling could have introduced selection bias. The cross-sectional design also precludes making predictive inferences. The use of a single-item health literacy measure could also limit the reliability of our findings. Lastly, patient portal use was measured through self-report rather than objectively capturing actual patient portal use, which could limit the validity of our findings.

Despite these limitations, the study still contributes to the existing knowledge on patient portals. Its findings on the socio-behavioral factors that influence patient portal use among community-dwelling older adults could benefit future researchers who are looking to improve the adoption of patient portal interventions.

Implications for Future Research and Nursing Practice

Considering the potential influence of social support on patient portal adoption in the older population, future researchers should consider including accommodations for the patients’ designated care partner in the implementation of their patient portals, such as inviting them to the orientation session or designing the portal in such a way that would enable patients to allow their care partners access to their health information from the care partners’ own portal accounts. Researchers should also take into consideration that not all older adults have access to social support. Incorporating a virtual assistant that would guide users on how to use the features of the portal could improve its adoption among independent older adults. Similarly, simplifying the navigation of patient portals by imitating how a telephone menu operates could improve its usability, especially among older adults who might be more familiar with engaging with their health care providers through the telephone. Designing the patient portal landing page like a telephone menu (with buttons or links for accessing test results, requesting prescription refills, scheduling appointments, and other common portal activities) could help older adults who might otherwise have difficulty navigating a typical website. Beyond patient portal design and adoption, future researchers should consider examining the actual impact of patient portal use on health outcomes. Cost effectiveness analyses should also be undertaken to determine whether establishing a patient portal, which requires a considerable investment, would lead to the desired health-related outcomes.

Aside from informing future research, findings from this study could inform current nursing practice. As members of the most trusted profession, nurses could help facilitate the adoption of patient portals by recommending them to their patients. Nurses could also show their patients how to access the portals and, subsequently, how to navigate them. Receiving the recommendation and support from their nurses could encourage older patients to start using patient portals.

Conclusion

Efforts to increase patient portal use among older adults require attention to multiple factors including current level of health literacy, activation level, comfort in using information technology, degree of social support, and opportunities for social influence within day-to-day life. The significant association between marital status and patient portal use underscores the important role of social support in the elderly population and may indicate the need to provide extra training and support to older individuals who are living on their own or have less social support. Just as financial incentives for meaningful use of EHR technology propelled the use of portals, health care stakeholders (i.e., providers and insurers) could be further incentivized to focus more on providing the social support needed, such as pairing target users with health coaches who could serve as a proxy for social support, to increase adoption and long-term utilization of patient portals. Promoting the use of patient portals among older adults, independent or otherwise, could be a means of empowering older adults to become more actively engaged in their health care and could potentially narrow the grey digital divide.

Contributor Information

Maan Isabella Cajita, University of Illinois at Chicago College of Nursing, 845 S Damen Avenue, Chicago, IL 60612.

Marci Lee Nilsen, University of Pittsburgh School of Nursing, 3500 Victoria Street, Pittsburgh, PA 15261.

Taya Irizarry, University of Pittsburgh School of Nursing, 3500 Victoria Street, Pittsburgh, PA 15261.

Judith A. Callan, University of Pittsburgh School of Nursing, 3500 Victoria Street, Pittsburgh, PA 15261.

Scott R. Beach, University of Pittsburgh Center for Social and Urban Research, 3343 Forbes Avenue, Pittsburgh, PA 15260.

Ellen Swartwout, O’Neil Center/GetWellNetwork, 7700 Old Georgetown Road, Bethesda, MD 20814.

Laurel Person Mecca, McMaster University School of Nursing, 1280 Main Street West, Hamilton, ON, Canada L8S 4K.

Richard Schulz, University of Pittsburgh Center for Social and Urban Research, 3343 Forbes Avenue, Pittsburgh, PA 15260.

Annette DeVito Dabbs, University of Pittsburgh School of Nursing, 3500 Victoria Street, Pittsburgh, PA 15261.

References

  1. Aljabri D, Dumitrascu A, Burton MC, White L, Khan M, Xirasagar S, . . . Naessens J (2018). Patient portal adoption and use by hospitalized cancer patients: a retrospective study of its impact on adverse events, utilization, and patient satisfaction. BMC Medical Informatics and Decision Making, 18(1), 70. doi: 10.1186/s12911-018-0644-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anderson M, & Perrin A (2017). Tech adoption climbs among older adults. Retrieved from http://www.pewinternet.org/2017/05/17/technology-use-among-seniors/
  3. Arcury TA, Quandt SA, Sandberg JC, Miller DP Jr., Latulipe C, Leng X, . . . Bertoni AG (2017). Patient portal utilization among ethnically diverse low income older adults: Observational study. JMIR Medical Informatics, 5(4), e47. doi: 10.2196/medinform.8026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brown SA, Venkatesh V, & Bala H (2006). Household technology use: Integrating household life cycle and the model of adoption of technology in households. The Information Society, 22(4), 205–218. doi: 10.1080/01972240600791333 [DOI] [Google Scholar]
  5. Chew LD, Bradley KA, & Boyko EJ (2004). Brief questions to identify patients with inadeqaute health literacy. Family Medicine, 36(8), 588–594. [PubMed] [Google Scholar]
  6. Chew LD, Griffin JM, Partin MR, Noorbaloochi S, Grill JP, Snyder A, . . . Vanryn M (2008). Validation of screening questions for limited health literacy in a large VA outpatient population. Journal of General Internal Medicine, 23(5), 561–566. doi: 10.1007/s11606-008-0520-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Conseil Européen pour la Recherche Nucléaire. (n.d.). The birth of the web. Retrieved from https://home.cern/science/computing/birth-web
  8. Couture ÉM, Chouinard M-C, Fortin M, & Hudon C (2018). The relationship between health literacy and patient activation among frequent users of healthcare services: a cross-sectional study. BMC Family Practice, 19(1). doi: 10.1186/s12875-018-0724-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Czaja SJ, Charness N, Fisk AD, Hertzog C, Nair SN, Rogers WA, & Sharit J (2006). Factors Predicting the Use of Technology: Findings from the Center for Research and Education on Aging and Technology Enhancement (CREATE). Psychology and Aging, 21(2), 333–352. doi: 10.1037/0882-7974.21.2.333] [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. DeSalvo KB, Fisher WP, Tran K, Bloser N, Merrill W, & Peabody J (2006). Assessing measurement properties of two single-item general health measures. Quality of Life Research, 15(2), 191–201. doi: 10.1007/s11136-005-0887-2 [DOI] [PubMed] [Google Scholar]
  11. Health Literacy: A Prescription to End Confusion. (2004). (Nielsen-Bohlman L, Panzer A, & Kindig D Eds.). Washington, DC: The National Academies Press. [PubMed] [Google Scholar]
  12. Hibbard JH (2017). Patient activation and health literacy: What’s the difference? how do each contribute to health outcomes. Studies in Health Technology and Informatics, 240, 251–262. doi: 10.3233/978-1-61499-790-0-251 [DOI] [PubMed] [Google Scholar]
  13. Hibbard JH, Mahoney ER, Stockard J, & Tusler M (2005). Development and testing of a short form of the patient activation measure. Health Services Research, 40(6), 1918–1930. doi: 10.1111/j.1475-6773.2005.00438.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hibbard JH, Stockard J, Mahoney ER, & Tusler M (2004). Development of the patient activation measure (PAM): Conceptualizing and measuring activation in patients and consumers. Health Services Research, 39(4), 1005–1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Irizarry T, DeVito Dabbs A, & Curran CR (2015). Patient portals and patient engagement: A state of the science review. Journal of Medical Internet Research, 17(6), e148. doi: 10.2196/jmir.4255 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Khamis H (2008). Measures of Association: How to Choose? Journal of Diagnostic Medical Sonography, 24(3), 155–162. doi: 10.1177/8756479308317006 [DOI] [Google Scholar]
  17. Latulipe C, Gatto A, Nguyen HT, Miller DP, Quandt SA, Bertoni AG, . . . Arcury TA (2015). Design considerations for patient portal adoption by low-income older adults. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2015, 3859–3868. doi: 10.1145/2702123.2702392 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Levy H, Janke AT, & Langa KM (2015). Health literacy and the digital divide among older Americans. Journal of General Internal Medicine, 30(3), 284–289. doi: 10.1007/s11606-014-3069-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Nahm E-S, Sagherian K, & Zhu S (2016). Use of patient portals in older adults: A comparison of three samples. Nursing Informatics, 225, 354–358. doi: 10.3233/978-1-61499-658-3-354 [DOI] [PubMed] [Google Scholar]
  20. Nahm E-S, Zhu S, Bellantoni M, Keldsen L, Charters K, Russomanno V, . . . Smith L (2018). Patient portal use among older adults: What is really happening nationwide? Journal of Applied Gerontology, 00(0), 1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Patel V, & Johnson C (2018). Individuals’ use of online medical records and technology for health needs. ONC Data Brief. https://www.healthit.gov/sites/default/files/page/2018-03/HINTS-2017-Consumer-Data-Brief-3.21.18.pdf
  22. Powell K (2017). Patient-perceived facilitators of and barriers to electronic portal use: A systematic review. CIN: Computers, Informatics, Nursing, 35(11), 565–573. doi: 10.1097/CIN.0000000000000377 [DOI] [PubMed] [Google Scholar]
  23. Powell K, & Myers C (2018). Electronic patient portals: Patient and provider perceptions. Online Journal of Nursing Informatics, 22(1), 1–19. [Google Scholar]
  24. Rigby M, Georgiou A, Hypponen H, Ammenwerth E, de Keizer N, Magrabi F, & Scott P (2015). Patient Portals as a Means of Information and Communication Technology Support to Patient- Centric Care Coordination - the Missing Evidence and the Challenges of Evaluation. A joint contribution of IMIA WG EVAL and EFMI WG EVAL. Yearb Med Inform, 10(1), 148–159. doi: 10.15265/IY-2015-007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Rosenzveig A, Kuspinar A, Daskalopoulou SS, & Mayo NE (2014). Toward patient-centered care: a systematic review of how to ask questions that matter to patients. Medicine (Baltimore), 93(22), e120. doi: 10.1097/MD.0000000000000120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Sakaguchi-Tang DK, Bosold AL, Choi YK, & Turner AM (2017). Patient portal use and experience among older adults: Systematic review. JMIR Medical Informatics, 5(4), e38. doi: 10.2196/medinform.8092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Schnock KO, Snyder JE, Fuller TE, Duckworth M, Grant M, Yoon C, . . . Dykes PC (2019). Acute Care Patient Portal Intervention: Portal Use and Patient Activation. J Med Internet Res, 21(7), e13336. doi: 10.2196/13336 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Skolasky RL, Green AF, Scharfstein D, Boult C, Reider L, & Wegener ST (2011). Psychometric properties of the patient activation measure among multimorbid older adults. Health Services Research, 46(2), 457–478. doi: 10.1111/j.1475-6773.2010.01210.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Smith SG, Curtis LM, Wardle J, von Wagner C, & Wolf MS (2013). Skill set or mind set? associations between health literacy, patient activation and health. PLoS One, 8(9). doi: 10.1371/journal.pone.0074373 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Smith SG, Pandit A, Rush SR, Wolf MS, & Simon C (2015). The association between patient activation and accessing online health information: Results from a national survey of US adults. Health Expectations, 18(6), 3262–3273. doi: 10.1111/hex.12316 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Solomon M, Wagner SL, & Goes J (2012). Effects of a web-based intervention for adults with chronic conditions on patient activation: Online randomized controlled trial. Journal of Medical Internet Research, 14(1), e32. doi: 10.2196/jmir.1924 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Taha J, Sharit J, & Czaja SJ (2014). The impact of numeracy ability and technology skills on older adults’ performance of health management tasks using a patient portal. Journal of Applied Gerontology, 33(4), 416–436. doi: 10.1177/0733464812447283 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Venkatesh V, Thong JY, & Xu X (2016). Unified theory of acceptance and use of technology: A synthesis and the road ahead. Journal of the Association for Information Systems, 17(5), 328–376. [Google Scholar]
  34. Zarcadoolas C, Vaughon WL, Czaja SJ, Levy J, & Rockoff ML (2013). Consumers’ perceptions of patient-accessible electronic medical records. Journal of Medical Internet Research, 15(8), e168. doi: 10.2196/jmir.2507 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES