Abstract
We examine Internet use and eHealth literacy among older adults (aged 55+ years) who were patients at clinics serving low-income populations. Participants included 200 minority and White adults who completed interviews based on a technology acceptance conceptual model. A total of 106 participants (53.0%) used the Internet; utilization was associated with personal characteristics (age, ethnicity, education, poverty), computer characteristics (number of e-devices, computer stress), social support (marital status), and health knowledge and attitudes (health literacy, medical decision making, health information sources), but not health status. Of the 106 participants who used the Internet, 52 (49.1%) had high eHealth literacy; eHealth literacy was associated with computer characteristics (number of e-devices, computer stress), and health knowledge and attitudes (medical decision making, health information sources). In multivariate analysis, computer stress maintained a significant inverse association with eHealth literacy. Educational interventions to help older adults successfully use technology and improve eHealth literacy must be identified.
Keywords: health self-management, health literacy, eHEALS, digital divide
Introduction
Older adults often have multiple chronic conditions and are major users of health care (Ferris et al., 2018). Supporting the health self-management for older adults with multiple chronic conditions is a central component of improving the overall health of this population (Liddy, Blazkho, & Mill, 2014). However, several barriers compromise individuals’ self-management of chronic conditions, such as lack of financial resources, poor access to medical resources, complexity of social support, and confusion due to contradictory information provided by multiple health care providers (Liddy et al., 2014). New electronic technology provides opportunities to help older adults address these limitations and improve their self-management of chronic conditions.
Although electronic technology has become nearly ubiquitous, older adults, particularly those age 75 years and older, still lag behind younger persons in its adoption, especially with regard to the Internet (Masterson-Creber et al., 2016). Commonly cited reasons for older adults’ lack of electronic technology uptake include attitude, awareness, inappropriate design, cost, lack of self-efficacy, and a general lack of interest (Delello & McWhorter, 2017).
The term “eHealth” refers to electronic technology applications in the health domain, in particular opportunities for health care provided via the Internet (Scantlebury, Booth, & Hanley, 2017). The Internet can be a resource to individuals seeking knowledge about health conditions and their treatments; therefore, it is not surprising that the Internet is considered a tool for transforming the health care industry (Scantlebury et al., 2017). The Internet has extensive readily available information, but the reliability and accuracy of that information is difficult to assess (Nguyen, Mosadeghi, & Almario, 2017). This is problematic as most online health seekers believe all or most information they find on the Internet and do not assess the sources of the website they visit (Morahan-Martin, 2004; Sbaffi & Rowley, 2017).
eHealth literacy is “the ability to seek, find, understand, and appraise health information from electronic sources and apply the knowledge gained to addressing or solving a health problem” (Norman & Skinner, 2006b, p. e9). Research examining the levels of eHealth literacy among older adults and factors associated with eHealth literacy levels is limited. Choi and DiNitto (2013a) examined factors associated with eHealth literacy among adults aged 60 years and older receiving home-delivered meals in central Texas, and found only 17% of adults age 60 years and older used the Internet. In addition, they found that affordability was the primary barrier to Internet use with the number of older adults who had discontinued Internet use due to costs equal to the number who were currently using it (Choi & DiNitto, 2013a). Furthermore, eHealth literacy was negatively associated with age; those of older ages had lower perceived eHealth self-efficacy (Choi & DiNitto, 2013a).
It is necessary to assess patients’ eHealth literacy and identify its determining factors to ensure they can use available eHealth resources effectively (Richtering et al., 2017). However, little research has examined the eHealth literacy of older adults, and the factors associated with eHealth literacy levels. This information is particularly important to address eHealth among the members of low-income and minority older adults who have had limited access to technology (Levy, Janke, & Langa, 2015). The eHealth Literacy Scale (eHEALS) is one of a few existing validated scales assessing eHealth literacy (Norman & Skinner, 2006a; Richtering et al., 2017). It includes eight items scored on a 5-point Likert-type scale and was developed to assess individuals’ perceptions about their skills using information technology to learn about health issues and evaluate the “fit” between eHealth programs and users (Norman & Skinner, 2006a).
Davis’s technology acceptance model (TAM; Davis, 1989) and the person-environmental interaction model (Lawton, 1989) together provide a conceptual framework for understanding technology adoption and proficiency among older adults (Adamson & Shine, 2003; Arcury et al., 2017; Bruner & Kumar, 2005; Chau, 1996; Darkins et al., 2008; Or & Karsh, 2009). Computer characteristics (including whether a person owns or uses electronic devices and stress in using these devices), social support, general health knowledge and attitudes, and health status shape eHealth technology adoption and proficiency, in conjunction with such personal characteristics as age, gender, ethnicity, education, employment, and poverty influence the degree to which older adults utilize personal health technology and their perception of its usability and usefulness (Venkatesh & Davis, 2000). The Internet is an important (perhaps necessary) means to utilize personal health technology, and eHealth literacy is an important indicator of effective personal health technology utilization.
This analysis examines the eHealth literacy levels of low-income and minority older adults and the factors associated with levels by addressing two aims. Among older adults (aged 55 years and older) who are patients at clinics serving low-income populations, the first aim is to determine Internet use, and to identify computer characteristics, social support, general health knowledge and attitudes, health status, and personal characteristics associated with Internet use. Among the older adults who use the Internet, the second aim is to identify the levels of eHealth literacy using the eHEALS scale, and to identify computer and Internet characteristics, social support, general health knowledge and attitudes, health status, and personal characteristics associated with eHealth literacy level.
Method
Participant Recruitment and Participation
The data for this analysis were collected as part of a larger study examining factors related to patient portal use among older patients (Arcury et al., 2017). We recruited participants from urban and rural clinics that primarily serve low-income patients and patients from minority communities. The urban clinic was the Outpatient Department (OPD) of the Wake Forest Baptist Health Internal Medicine residency program, located in Winston-Salem, North Carolina (NC). OPD serves ethnically diverse, low-income, predominantly Medicare and Medicaid patients. Two members of Community Partners HealthNet were the rural clinics. Community Partners HealthNet is a Health Center Controlled Network (established in 1999) to implement practice management systems for community health centers. Greene County Health Care, Inc., and West Caldwell Health Council, Inc., serve the rural areas of Greene and Caldwell Counties, NC, respectively. Greene County Health Care, Inc., has six clinic locations. West Caldwell Health Council, Inc., has two clinic locations.
Participants included community-dwelling adults aged 55 years and older, who were being treated for a chronic disease (diabetes, hypertension, dyslipidemia, or cardiovascular disease), who spoke English or Spanish, and were in sufficiently good health to give informed consent and complete an interview. The sample included those 55 years and older because familiarity with and use of electronic media varies greatly by age among older adults starting at age 55 years. We recruited a majority of participants using a three-step process. First, with the assistance of clinic staff and physicians, we generated lists of patients who met the inclusion criteria. Second, we randomly selected potential participants from these lists and sent letters introducing and describing the study. Third, we made follow-up phone calls to describe the study and to schedule interviews with those who were sent the letters. In addition, we recruited Spanish-speaking participants as they came to one set of rural clinics. The clinics often do not have accurate telephone information for these participants due to the frequency with which they change telephone service. We approached individuals fitting the inclusion criteria described the study and scheduled interviews for a later date. The study protocol was approved by the investigators’ Institutional Review Board, and all participants provided signed informed consent.
Participants included 200 minority (African American, Latino, American Indian, Asian) and White older adult patients who completed baseline interviews. Data collectors attempted telephone contacts with 628 patients who were sent a letter or who were contacted in a clinic. Of the 628 attempted telephone contacts, 110 had a nonworking telephone number, 111 could not be contacted by telephone, 13 were deceased, and 394 were contacted for a contact rate of 62.7%. Of the 394 who were contacted, 194 refused to participate, for a refusal rate of 49.2%, and 200 participants were successfully enrolled and completed interviews, for an overall participation rate of 31.8%. Those who refused to participate were equally divided among women and men. However, more White (42.2%) than African American (22.1%), Latino (0%), American Indian (0%), or Asian (0%) patients refused to participate, and more urban clinic (65.9%) than rural clinic (28.8%) patients refused to participate.
Data Collection
The questionnaire included items eliciting information on personal characteristics, computer characteristics, social support, health knowledge and attitudes, health status, and eHealth literacy. We used existing items and scales when available in developing the questionnaire.
Trained interviewers administered the questionnaires in person, usually at the participants’ homes or at the clinic where they received medical care, from November 2014 through May 2016. The interview generally took 1 hr to complete and ranged in length from 45 min to 2 hr. We gave participants an incentive of US$20 for completing the interview. REDCap (Research Electronic Data Capture), a secure web-based system, was used to record interview data (Harris et al., 2009).
Measures
We measured eHealth literacy with the eight-item eHEALS (Chung & Nahm, 2015; Norman & Skinner, 2006a). Among the 106 participants who used the Internet, this scale had a Cronbach’s alpha of .73. Because the scores were not normally distributed, they were split at the mean, with high eHealth literacy defined as a score of 29 or greater on the eHEALS, and low eHealth literacy defined as a score of less than 29 on the eHEALS.
We included the participant personal characteristics such as age (in the categories 55-59 years, 60-64 years, 65-69 years, and 70 years and older), gender, ethnicity (White and minority), education (high school or less, and greater than high school), and employment (not employed vs. employed part-time or full-time) in the analysis. Poverty level, defined by the United States Federal Government and used to qualify for assistance programs, was based on the total household income adjusted for the total number of residents and for the year in which the data were collected (U.S. Census Bureau, 2017). Poverty level was placed in the categories less than or equal to 200% of poverty level and more than 200% of poverty level.
We included three computer characteristic measures. Number of e-devices at home included the number of desktop and laptop computers, tablets, and smartphones participants reported having in their homes; values are reported in the categories of 0, 1, and 2 or more. Frequency of Internet use had the values of less than once a day and at least once a day. Stress experienced when using a computer (Yagil, Cohen, & Beer, 2016) had the values of any stress (low stress, moderate stress, much stress, or very much stress) and no stress.
We included three social support measures. Marital status had the values currently married and not currently married. Having a care partner was dichotomous. A care partner was defined for the participants as
someone who helps you with activities and questions about your health. These activities and questions include simple things, like reminding you to take your prescription and about an upcoming doctor’s appointment, or finding information about something the doctor has told you; they can include more substantial assistance, like taking you to an appointment, helping you take your medicine, and helping you exercise; and they can include personal assistance, like helping you get dressed and bathe. Those who help you can be your spouse, brothers or sisters, children, or friends. The person who helps you may live with you, but they may also live in another house. They might even live in another town or city and help you by phone or the internet.
Household structure had the values single person, two persons, or more than two-person household.
We included three measures of health knowledge and attitudes. The Newest Vital Sign was used to differentiate those with adequate versus inadequate general health literacy (Weiss et al., 2005). The scale had a range of 0 to 6, and had a Cronbach’s alpha of .82. Based on the recommendation of the scale developers, those with a score of 4 to 6 were considered to have adequate general health literacy, and those with a score of 0 to 3 were considered to have inadequate general health literacy. Medical decision-making attitude was based on a dichotomous variable about participants’ preference to rely on doctor’s knowledge when making health-related decisions. The number of health information sources was calculated by asking what sources participants used to find information about a health condition including, but not limited to, the Internet, medical books and magazines, television programs, print literature in medical offices, and talking with health care providers.
We measured health status with the SF-12v2® (Ware, Krosinski, & Keller, 1996) and the Charlson Comorbidity Index (Charlson, Pompei, Ales, & MacKenzie, 1987). The SF-12v2® assesses general health-related quality of life and includes a mental component score (MCS) and physical component score (PCS). MCS and PCS were scored and standardized using Health Outcomes™ Scoring Software, Version 4.5 (2011). The Charlson Comorbidity Index is the sum of 18 different self-reported chronic conditions.
Analysis
All analyses were performed using SAS 9.4 (Cary, NC). Participants’ personal characteristics, computer characteristics, social support, health knowledge and attitudes, and health status were described as frequency and percentages for categorical variables, and mean and standard deviation (SD) for continuous variables for the total sample, and were also compared between patients who had Internet use and those who had no Internet use, using chi-square tests for categorical variables and student’s t tests for continuous variable(s). Similarly, comparisons were also made between patients who had high eHealth literacy and those who had low eHealth literacy. A multivariate logistic regression model was used to examine association between ethnicity, number of e-devices, frequency of Internet use, computer stress, general health literacy, medical decision-making attitude (doctor’s knowledge) and number of health information sources, and high eHealth literacy. A characteristic was included in the logistic regression model if it had a statistically significance associations at 0.10 level with high eHealth literacy based on bivariate analyses. Odds ratios with the corresponding 95% confidence interval were estimated for each characteristic. The results of a sensitivity analysis for predicting eHealth literacy using eHealth literacy as a continuous measure are reported in the supplementary table. All tests were two-sided and performed at significance level of 0.05.
Results
Participant Characteristics, Computer Characteristics, Social Support, Health Knowledge and Attitudes, and Health Status
About one quarter (27.5%) of participants were aged 55 to 59 years, with 31.5% aged 60 to 64 years, 26.0% aged 65 to 69 years, and 15.0% were aged 70 years or older (Table 1). More participants were female (58.0%) than male (42.0%). Most (60.0%) were minority, 35.0% had greater than a high school education, and 80.0% were not employed. Most (83.3%) had household incomes at 200% of poverty or less.
Table 1.
Participant Characteristics for Total Sample (n = 200).
| Participant characteristics | n | % |
|---|---|---|
| Age in years | ||
| 55-59 | 55 | 27.5 |
| 60-64 | 63 | 31.5 |
| 65-69 | 52 | 26.0 |
| 70 and older | 30 | 15.0 |
| Gender | ||
| Male | 84 | 42.0 |
| Female | 116 | 58.0 |
| Ethnicity | ||
| White | 80 | 40.0 |
| Minoritya | 120 | 60.0 |
| Education | ||
| High school or less | 130 | 65.0 |
| Greater than high school | 70 | 35.0 |
| Employment | ||
| Not employed | 160 | 80.0 |
| Employed | 40 | 20.0 |
| Poverty level | ||
| 200% of poverty or less | 160 | 83.3 |
| Greater than 200% of poverty level | 32 | 16.7 |
Participants include 90 African American, 26 Latino, two American Indian, and one Asian.
More than one quarter (28.0%) of participants did not own an e-device, with 47.5% owning two or more (Table 2). Most participants (69.5%) used the Internet less than once each day, and 75.9% experienced stress when using a computer. Fewer than half (42.0%) were married, but 62.0% had a care provider. Thirty-five percent lived alone.
Table 2.
Participant Computer Characteristics, Social Support, Health Knowledge and Attitudes, and Health Status for Total Sample (n = 200).
| n | % | |
|---|---|---|
| Computer and Internet characteristics | ||
| Number of e-devices | ||
| 0 | 56 | 28.0 |
| 1 | 49 | 24.5 |
| 2 or more | 95 | 47.5 |
| Frequency of Internet use | ||
| Less than once a day | 139 | 69.5 |
| At least once a day | 61 | 30.5 |
| Computer stress | ||
| Any stress | 151 | 75.9 |
| No stress | 48 | 24.1 |
| Social support | ||
| Marital status | ||
| Married | 84 | 42.0 |
| Not married | 116 | 58.0 |
| Care partner | ||
| Yes | 124 | 62.0 |
| No | 76 | 38.0 |
| Household structure | ||
| 1 person | 70 | 35.0 |
| 2 persons | 78 | 39.0 |
| 3 or more persons | 52 | 26.0 |
| Health knowledge and attitudes | ||
| General health literacy | ||
| Inadequate | 136 | 72.0 |
| Adequate | 53 | 28.0 |
| Medical decision-making attitude | ||
| Prefer to rely on doctor’s knowledge | 112 | 56.9 |
| Do not prefer to rely on doctor’s knowledge | 85 | 43.1 |
| M | SD | |
| Health knowledge and attitudes | ||
| Number of health information sources | 3.0 | 1.7 |
| Health status | ||
| SF-12 | ||
| Mental component score | 50.2 | 12.3 |
| Physical component score | 37.6 | 12.2 |
| Modified Charlson Index | 5.9 | 2.7 |
Almost three quarters (72.0%) had inadequate general health literacy, and 56.9% preferred to rely on the doctor’s knowledge when making medical decisions. The mean of health information sources used by the participants was 3 (SD = 1.7). The mean for the MCS was 50.2 (SD = 12.3), and for the PCS was 37.6 (SD = 12.2). The mean for the Modified Charlson Index was 5.9 (SD = 2.7).
Factors Associated With Internet Use
A little more than half (53%) of the participants used the Internet (Table 3). A greater percentage of those less than age 65 years (65.5% of those aged 55-59 years, and 61.9% of those aged 60-64 years) used the Internet than those aged 65 years or older (42.3% of those aged 65-69 years, 30% of those aged 70 years and older). Internet use did not differ by gender. A greater percentage of White participants (67.5%) than minority participants (43.3%) used the Internet; comparison of African American and Latino participants showed no statistically significant difference (results not shown). Those with greater than a high school education (75.7%) were more likely to use the Internet than those with high school education or less (40.8%). Employment was not related to Internet use. Those with an income greater than 200% of poverty (75.0%) were more likely to use the Internet than those with a lower income (50.6%).
Table 3.
Associations of Personal Characteristics With Internet Use and High eHealth Literacy.
| Internet users |
High eHealth literacy |
|||||
|---|---|---|---|---|---|---|
|
n = 200 |
n = 106 |
|||||
| Personal characteristics | n | % | p valuea | n | % | p valuea |
| Total | 106 | 53.0 | — | 52 | 49.1 | — |
| Age in years | <.01 | .40 | ||||
| 55-59 | 36 | 65.5 | 14 | 38.9 | ||
| 60-64 | 39 | 61.9 | 22 | 57.9 | ||
| 65-69 | 22 | 42.3 | 12 | 54.5 | ||
| 70 and older | 9 | 30.0 | 4 | 44.4 | ||
| Gender | .67 | .61 | ||||
| Male | 46 | 54.7 | 21 | 46.7 | ||
| Female | 60 | 51.7 | 31 | 51.7 | ||
| Ethnicity | <.01 | .10 | ||||
| White | 54 | 67.5 | 31 | 57.4 | ||
| Minority | 52 | 43.3 | 21 | 41.2 | ||
| Education | <.01 | .14 | ||||
| High school or less | 53 | 40.8 | 22 | 42.3 | ||
| Greater than high school | 53 | 75.7 | 30 | 56.6 | ||
| Employment | .18 | .53 | ||||
| Not employed | 81 | 50.6 | 41 | 51.3 | ||
| Employed | 25 | 62.5 | 11 | 44.0 | ||
| Poverty level | .01 | .16 | ||||
| 200% of poverty or less | 81 | 50.6 | 37 | 46.3 | ||
| Greater than 200% of poverty level | 24 | 75.0 | 15 | 62.5 | ||
chi-square test.
Computer characteristics were associated with Internet use (Table 4). Those with two or more e-devices in their homes were much more likely to use the Internet (82.1%) than those with one (25.5%) or no home e-devices (1.8%). Those experiencing any stress while using a computer (41.1%) were less likely to use the Internet than those who experienced no stress (91.7%). A greater percentage of those who were married (64.3%) used the Internet than those who were not married (44.8%). Having a care partner and household structure was not related to Internet use.
Table 4.
Associations of Computer Characteristics, Social Support, and Health Knowledge and Attitudes With Internet Use and High eHealth Literacy.
| Internet users n = 200 |
p valuesa |
High eHealth literacy n = 106 |
p valuea |
|||
|---|---|---|---|---|---|---|
| n | % | n | % | |||
| Computer characteristics | ||||||
| Number of e-devices | <.01 | .03 | ||||
| 0 | 1 | 1.8 | 0 | — | ||
| 1 | 27 | 25.5 | 9 | 33.3 | ||
| 2 or more | 78 | 82.1 | 43 | 46.7 | ||
| Frequency of Internet use | .06 | |||||
| Less than once a day | — | 17 | 38.6 | |||
| At least once a day | — | 35 | 57.4 | |||
| Computer stress | <.01 | <.01 | ||||
| Any stress | 62 | 41.1 | 22 | 36.1 | ||
| No stress | 44 | 91.7 | 30 | 68.2 | ||
| Social support | ||||||
| Marital status | <.01 | .38 | ||||
| Married | 54 | 64.3 | 24 | 45.3 | ||
| Not married | 52 | 44.8 | 28 | 53.8 | ||
| Care partner | .09 | .48 | ||||
| Yes | 60 | 48.4 | 31 | 52.5 | ||
| No | 46 | 60.5 | 21 | 45.7 | ||
| Household structure | .08 | .53 | ||||
| 1 person | 32 | 45.7 | 18 | 56.3 | ||
| 2 persons | 49 | 62.8 | 21 | 43.8 | ||
| 3 or more persons | 25 | 48.1 | 13 | 52.0 | ||
| Health knowledge and attitudes | ||||||
| General health literacy | <.01 | .09 | ||||
| Inadequate | 58 | 42.6 | 24 | 42.1 | ||
| Adequate | 44 | 83.0 | 26 | 59.1 | ||
| Medical decision-making attitude | <.01 | .04 | ||||
| Prefer to rely on doctor’s knowledge | 48 | 42.9 | 18 | 38.3 | ||
| Do not prefer to rely on doctor’s knowledge | 58 | 68.2 | 34 | 58.6 | ||
chi-square test.
Fewer of those with inadequate general health literacy (42.6%) than those with adequate general health literacy (83.0%) used the Internet. Those who preferred to rely on their doctor’s knowledge for medical decision making (42.9%) were less likely to use the Internet than those who did not prefer to rely on their doctor’s knowledge (68.2%). Those who used the Internet had an average of 3.4 (SD = 1.5) health information sources; those who did not use the Internet used an average of 2.6 (SD = 1.7) health information sources (Table 5). Health status, as measured by the SF-12 MCS subscale and PCS subscale, and the Charlson Index, was not associated with Internet use.
Table 5.
Associations of Health Status With Internet Use and eHealth Literacy.
| No |
Yes |
||||
|---|---|---|---|---|---|
| Internet use (n = 200) | M | SD | M | SD | p valuea |
| Health knowledge and attitudes | |||||
| Number of health information sources | 2.6 | 1.7 | 3.4 | 1.5 | <.01 |
| Health status | |||||
| SF-12 | |||||
| MCS | 50.8 | 13.4 | 49.7 | 11.4 | .55 |
| PCS | 37.4 | 11.9 | 37.8 | 12.5 | .84 |
| Modified Charlson Index | 5.8 | 2.9 | 6.1 | 2.6 | .37 |
| Low |
High |
||||
| eHealth literacy (n = 106) | M | SD | M | SD | p valuea |
| Health knowledge and attitudes | |||||
| Number of health information sources | 3.0 | 1.5 | 3.8 | 1.4 | <.01 |
| Health status | |||||
| SF-12 | |||||
| MCS | 48.1 | 12.2 | 51.2 | 10.4 | .17 |
| PCS | 37.0 | 13.3 | 38.5 | 11.7 | .53 |
| Modified Charlson Index | 5.8 | 2.5 | 6.4 | 2.7 | .19 |
Note. MCS = mental component score; PCS = physical component score.
Two-sample t test.
Factors Associated With eHealth Literacy
Among the 106 participants who used the Internet, eHealth literacy scores had a range of 8 to 40, and an overall mean of 28.4 (SD = 7.1). Fifty-two participants (49.1%) had high eHealth literacy, and 54 (50.9%) had low eHealth literacy. Having high eHealth literacy did not differ in terms of personal characteristics among older adults who use the Internet (Table 3). However, having high eHealth literacy was associated with computer characteristics (Table 4). A greater percentage of those with two or more e-devices in their homes (46.7%) had high eHealth literacy than those with one e-device (33.3%) or no e-device (0). More of those experiencing no stress while using a computer (68.2%) had high eHealth literacy, compared with those who experienced any stress (36.1%). None of the social support measures were associated with high eHealth literacy.
Fewer of those who preferred to rely on their doctor’s knowledge for medical decision making (38.3%) had high eHealth literacy than those who did not prefer to rely on their doctor’s knowledge (58.6%). Those with high eHealth literacy used an average of 3.8 (SD = 1.45) health information sources; those who did not have high eHealth literacy used an average of 3.0 (SD = 1.5) health information sources. Health status, as measured by the SF-12 MCS subscale and PCS subscale, and the Charlson Index, were not associated with high eHealth literacy.
A multivariate model of the measures with a statistically significant bivariate association with high eHealth literacy shows that no computer stress was the one measure that remained significantly associated (Table 6). The odds ratio of no computer stress with high eHealth literacy was 3.05, with a 95% confidence interval of [1.13, 8.23], and ap value of .03. In addition, the number of health information sources had a marginal positive association with having a high eHealth literacy score. The odds ratio of health information sources with high eHealth literacy was 1.41, with a 95% confidence interval of [1.00, 1.99], and a p value of .053. A second multivariate model in which all measures with a p value < .20 in the bivariate analyses was also calculated (results not shown). This second model did not differ from the first; only no computer stress was significantly associated with having high eHealth literacy, and the association of number of health information sources remained marginally associated with high eHealth literacy.
Table 6.
Multivariate Logistic Regression Analysis Results for Predicting High eHealth Literacy Among Internet Users (n = 101).
| Odds ratio | 95% confidence limits | p value | |
|---|---|---|---|
| Personal characteristics | |||
| Ethnicity: White (vs. minority) | 1.18 | [0.40, 3.47] | .77 |
| Computer characteristics | |||
| Number of e-devices: 2 or more (vs. 1) | 2.04 | [0.67, 6.23] | .21 |
| Frequency of Internet use: ⩾ 1/day (vs. < 1/day) | 1.06 | [0.37, 3.03] | .91 |
| No computer stress (vs. any stress) | |||
| Health knowledge and attitudes | 3.05 | [1.13, 8.23] | .03 |
| General health literacy: Adequate (vs. inadequate) | 1.12 | [0.39, 3.25] | .84 |
| Prefer to rely on doctor’s knowledge for medical decision making | 0.62 | [0.24, 1.62] | .33 |
| Number of ways tries to find information about health conditions | 1.41 | [0.99, 1.99] | .053 |
Discussion
Little more than one half of the older adults participating in this study use the Internet, and about one half of those who use the Internet score high on the eHEALS, a measure of eHealth literacy. Therefore, one quarter of all the participants have high eHealth literacy.
The level of general Internet utilization among these older adult participants in terms of age, education, and income is not surprising for the patients who receive care at clinics focused on care for low-income populations relative to national surveys (Keenam, 2009; Pew Research Center, 2014). Other analyses find somewhat higher Internet use among adults age 65 years and older, 60% by AARP (Keenam, 2009) and 59% by the Pew Research Center (2014) compared with 53% among adults age 55 years and older in our study. However, our percentage of Internet use reflects the high percentage of low-income participants in our study compared with these national surveys. Our analysis finds lower general Internet use among older adults with increasing age, lower educational attainment, and minority ethnic status. These results are similar to those of other recent studies (Choi & DiNitto, 2013b; Hong & Cho, 2017; Vroman, Arthanat, & Lysack, 2015). We do not find gender differences in Internet utilization; gender differences are not consistent in the literature (Choi & DiNitto, 2013b; Kim, Lee, Christensen, & Merighi, 2016).
Our conceptual framework (Arcury et al., 2017) suggests several factors that should be considered in understanding the Internet utilization of older adults in addition to demographic characteristics, including computer characteristics, social support, health knowledge and attitudes, and health status. We find that general Internet utilization among older adults is associated with computer characteristics (number of e-devices, computer stress), social support in terms of being married, and health knowledge and attitudes (general health literacy, participation in decision making, number of health information sources), but not with health status. Several of these factors are similar to other research results. In terms of computer characteristics, Vroman et al. (2015) report that those older adults not using the Internet reported feeling intimidated and anxious with technology. In terms of social support, Vroman et al. (2015) and Choi and DiNitto (2013b) found that those living with a spouse were more likely to use the Internet. Choi and DiNitto (2013b) report that Internet usage was associated with having greater social capital, with Kim and colleagues (2016) reporting that digital utilization was associated with social engagement of older adults, particularly among women. We do not find health status to be associated with Internet utilization. This result differs from analyses that find that health status is directly associated with general digital technology use (Berkowsky, Rikard, & Cotten, 2015; Choi & DiNitto, 2013b; Hong & Cho, 2017; Levine, Lipsitz, & Linder, 2017). This difference may reflect our sample design, which required that participants had to be treated for one of four chronic diseases (diabetes, hypertension, dyslipidemia, or cardiovascular disease).
Our older adult participants who use the Internet do not vary in eHealth literacy by such personal characteristics as age, gender, ethnicity, education, or poverty. This differs from other analyses that report that these personal characteristics are related to eHealth literacy (Choi & DiNitto, 2013a; Tennant et al., 2015); however, by limiting our analysis of eHealth literacy to those using the Internet, we account for these personal characteristic differences. Other factors included in our conceptual model are related to eHealth literacy among the older adult Internet users. Although neither social support nor health status measures are related to eHealth literacy, computer characteristics (number of devices, Internet utilization frequency, computer stress), and health knowledge and attitudes (shared medical decision making, number of health information sources) have positive associations with eHealth literacy. In particular, the multivariate model indicates that lack of computer stress is a key to eHealth literacy. This is similar to other analyses that indicate that those older adults with more technology knowledge and trust are more eHealth literate (Gordon & Hornbrook, 2018; Paige, Krieger, & Stellefson, 2017; Tennant et al., 2015).
This research should be evaluated within its limitations. The sample was drawn from patients receiving care at three sets of clinics (one urban, two rural), and the participation rate is limited. These factors limit the generalizability of the results. At the same time, this survey did recruit a large, multiethnic, low-income sample that included both rural and urban patients. The measure of eHealth literacy, the eHEALS, is a validated scale that is becoming widely used (Chung & Nahm, 2015; Norman & Skinner, 2006a).
Conclusion
Increasing older adult Internet use can improve access to health information and improve health self-management (Delello & McWhorter, 2017; Masterson-Creber, Hickley, & Mauer, 2016). It is essential that older adults have the eHealth literacy to understand and determine the veracity of information that they find through Internet and other digital sources (Norman & Skinner, 2006b). Determining personal characteristics that are associated with limited Internet use and eHealth literacy among older adults is important to understanding their prevalence, but these personal characteristics do not provide leverage points to improve use or literacy. This conceptually based analysis (Arcury et al., 2017) indicates that improving computer skills and limiting computer stress may be the keys to improving older adults’ eHealth use and understanding. Cotten (2017) argues that we must identify ways to help older adults successfully use technologies, but that current design impedes this use. At the same time, in their review of eHealth literacy intervention, Watkins and Xie (2014) conclude that existing interventions are neither theory based nor use high-quality research design. Future research and intervention must continue to delineate leverage points for improving technology use and eHealth literacy among older adults, and conceptually based interventions that use these leverage points should be developed and tested.
Supplementary Material
Acknowledgments
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was supported by grant R01 HS021679 from the Agency for Health care Research and Quality.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Human Subjects Approval
This research was reviewed and approved by the Wake Forest University Health Sciences Institutional Review Board (FWA00001435) as Human Protocol IRB00024562.
Supplemental Material
Supplemental material for this article is available online.
References
- Adamson I, & Shine J (2003). Extending the new technology acceptance model to measure the end user information systems satisfaction in a mandatory environment: A bank’s treasury. Technology Analysis & Strategic Management, 15, 441–455. doi: 10.1080/095373203000136033 [DOI] [Google Scholar]
- 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]
- Berkowsky RW, Rikard RV, & Cotten SR (2015). Signing off: Predicting discontinued ICT usage among older adults in assisted and independent living, a survival analysis In Human aspects of IT for the aged population. Design for everyday life (pp. 389–398). Cham, Switzerland: Springer International Publishing. doi: 10.1007/978-3-319-20913-5_36 [DOI] [Google Scholar]
- Bruner GC, & Kumar A (2005). Explaining consumer acceptance of handheld Internet devices. Journal of Business Research, 58, 553–558. [Google Scholar]
- Charlson ME, Pompei P, Ales KL, & MacKenzie CR (1987). A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. Journal of Chronic Disease, 40, 373–383. [DOI] [PubMed] [Google Scholar]
- Chau PY (1996). An empirical investigation on factors affecting the acceptance of CASE by systems developers. Information and Management, 30, 269–280. [Google Scholar]
- Choi NG, & DiNitto DM (2013a). The digital divide among low-income homebound older adults: Internet use patterns, eHealth literacy, and attitudes toward computer/internet use. Journal of Medical Internet Research, 15(5), e93. doi: 10.2196/jmir.2645 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Choi NG, & DiNitto DM (2013b). Internet use among older adults: Association with health needs, psychological capital, and social capital. Journal of Medical Internet Research, 15(5), e97. doi: 10.2196/jmir.2333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chung SY, & Nahm ES (2015). Testing reliability and validity of the eHealth Literacy Scale (eHEALS) for older adults recruited online. Computers, Informatics, Nursing, 33, 150–156. doi: 10.1097/CIN.0000000000000146 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cotton SR (2017). Examining the roles of technology in aging and quality of life. J Gerontol B Psychol Sci Soci, 72, 823–826. doi: 10.1093/geronb/gbx109 [DOI] [PubMed] [Google Scholar]
- Darkins A, Ryan P, Kobb R, Foster L, Edmonson E, Wakefield B, & Lancaster AE (2008). Care coordination/home telehealth: The systematic implementation of health informatics, home telehealth, and disease management to support the care of veteran patients with chronic conditions. Telemedicine Journal and e-Health, 14, 1118–1126. doi: 10.1089/tmj.2008.0021 [DOI] [PubMed] [Google Scholar]
- Davis FD (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13, 319–340. [Google Scholar]
- Delello JA, & McWhorter RR (2017). Reducing the digital divide: Connecting older adults to iPad technology. Journal of Applied Gerontology, 36, 3–28. [DOI] [PubMed] [Google Scholar]
- Ferris R, Blaum C, Kiwak E, Austin J, Esterson J, Harkless G, … Tinetti ME (2018). Perspectives of patients, clinicians, and health system leaders needed to improve health care and outcomes of older adults with multiple chronic conditions. Journal of Aging and Health, February, 30, 778–799. doi: 10.1177/0898264317691166 [DOI] [PubMed] [Google Scholar]
- Gordon NP, & Hornbrook MC (2018). Older adults’ readiness to engage with eHealth patient education and self-care resources: A cross-sectional survey. BMC Health Services Research, 18(1), Article 220. doi: 10.1186/s12913-018-2986-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, & Conde JG (2009). Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational informatics support. Journal of Biomedical Informatics, 42, 377–381. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Health Outcomes™ Scoring Software [Computer program]. (2011). Version 4.5. Lincoln, RI: Optuminsight Life Sciences. [Google Scholar]
- Hong YA, & Cho J (2017). Has the digital health divide widened? Trends of health-related internet use among older adults from 2003 to 2011. Journal of Gerontology, Series B: Psychological Science & Social Science, 72, 856–863. doi: 10.1093/geronb/gbw100 [DOI] [PubMed] [Google Scholar]
- Keenam TA (2009). Internet use among midlife and older adults: An AARP Bulletin Poll. Washington, DC: AARP Knowledge Management; Available from https://www.aarp.org/technology/innovations/info-12-2009/bulletin_internet_09.html [Google Scholar]
- Kim J, Lee HY, Christensen MC, & Merighi JR (2016). Technology access and use, and their associations with social engagement among older adults: Do women and men differ? Journal of Gerontology, Series B: Psychological Science & Social Science, 72, 836–845. doi: 10.1093/geronb/gbw123 [DOI] [PubMed] [Google Scholar]
- Lawton MP (1989). Behavior-relevant ecological factors In Schaie KW & Schooler C (Eds.), Social structure and aging: Psychological processes (pp. 57–78). Hillsdale, NJ: Lawrence Erlbaum. [Google Scholar]
- Levine DM, Lipsitz SR, & Linder JA (2017). Changes in everyday and digital health technology use among seniors in declining health. Journal of Gerontology, Series A: Biological Science & Medical Science, 73, 552–559. doi: 10.1093/gerona/glx116 [DOI] [PubMed] [Google Scholar]
- Levy H, Janke AT, & Langa KM (2015). Health literacy and the digital divide among older Americans. Journal of General Internal Medicine, 30, 284–289. doi: 10.1007/s11606-014-3069-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liddy C, Blazkho V, & Mill K (2014). Challenges of self-management when living with multiple chronic conditions: Systematic review of the qualitative literature. Canadian Family Physician, 60, 1123–1133. [PMC free article] [PubMed] [Google Scholar]
- Masterson-Creber RM, Hickley KT, & Mauer MS (2016). Gerontechnologies for older patients with heart failure: What is the role of smartphones, tablets, and remote monitoring devices in improving symptom monitoring and self-care management? Current Cardiovascular Risk Reports, 10(10). doi: 10.1007/s12170-016-0511-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Masterson-Creber RM, Hickley KT, Mauer MS, Reading M, Hiraldo G, Hickey KT, & Iribarren S (2016). Review and analysis of existing mobile phone apps to support heart failure symptom monitoring and self-care management. Using the Mobile Application Rating Scale (MARS). JMIR mHealth and uHealth, 4(2), e74. doi: 10.2196/mhealth.5882 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Morahan-Martin J (2004). How internet users find, evaluate, and use online health information: A cross-cultural review. Journal of CyberPsychology & Behavior, 7, 497–510. [DOI] [PubMed] [Google Scholar]
- Nguyen A, Mosadeghi S, & Almario CV (2017). Persistent digital divide in access to and use of the internet as a resource for health information: Results from a California population-based study. International Journal of Medical Informatics, 103, 49–54. doi: 10.1016/j.ijmedinf.2017.04.008 [DOI] [PubMed] [Google Scholar]
- Norman CD, & Skinner HA (2006a). eHEALS: The eHealth Literacy Scale. Journal of Medical Internet Research, 8(4), e27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Norman CD, & Skinner HA (2006b). eHealth literacy: Essential skills for consumer health in a networked world. Journal of Medical Internet Research, 8(2), e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Or CK, & Karsh BT (2009). A systematic review of patient acceptance of consumer health information technology. Journal of American Informatics Association, 16, 550–560. doi: 10.1197/jamia.M2888 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Paige SR, Krieger JL, & Stellefson ML (2017). The influence of eHealth literacy on perceived trust in online health communication channels and sources. Journal of Health Communication, 22, 3–65. doi: 10.1080/10810730.2016.1250846 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pew Research Center. (2014). Older adults and technology use. Retrieved from http://www.pewinternet.org/2014/04/03/older-adults-and-technology-use/
- Richtering SS, Hyun K, Neubeck L, Coorey G, Chalmers J, Usherwood T, … Redfern J (2017). eHealth literacy: Predictors in a population with moderate-to-high cardiovascular risk. JIMR: Human Factors, 4(1), e4. doi: 10.2196/human-factors.6217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sbaffi L, & Rowley J (2017). Trust and credibility in web-based health information: A review and agenda for future research. Journal of Medical Internet Research, 9(6), e218. doi: 10.2196/jmir.7579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Scantlebury A, Booth A, & Hanley B (2017). Experiences, practices and barriers to accessing health information: A qualitative study. International Journal of Medical Informatics, 103, 103–108. [DOI] [PubMed] [Google Scholar]
- Tennant B, Stellefson M, Dodd V, Chaney B, Chaney D, Paige S, & Alber J (2015). eHealth literacy and Web 2.0 health information seeking behaviors among baby boomers and older adults. Journal of Medical Internet Research, 17(3), e70. doi: 10.2196/jmir.3992 [DOI] [PMC free article] [PubMed] [Google Scholar]
- U.S. Census Bureau. (2017). Poverty. Retrieved from http://www.census.gov/topics/income-poverty/poverty.html
- Venkatesh V, & Davis SS (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Sciences, 46, 186–204. [Google Scholar]
- Vroman KG, Arthanat S, & Lysack C (2015). “Who over 65 is online?” Older adults’ dispositions toward information communication technology. Computers in Human Behavior, 43, 156–166. doi: 10.1016/j.chb.2014.10.018 [DOI] [Google Scholar]
- Ware J, Krosinski M, & Keller SD (1996). A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care, 34, 220–233. [DOI] [PubMed] [Google Scholar]
- Watkins I, & Xie B (2014). eHealth literacy interventions for older adults: A systematic review of the literature. Journal of Medical Internet Research, 16(11), e225. doi: 10.2196/jmir.3318 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weiss BD, Mays MZ, Martz W, Castro KM, Pigone MP, … Mockabee FA (2005). Quick assessment of literacy programs in primary care: The Newest Vital Sign. Annals of Family Medicine, 3, 514–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yagil D, Cohen M, & Beer JD (2016). Older adults’ coping with the stress involved in the use of everyday technologies. Journal of Applied Gerontology, 35, 131–149. doi: 10.1177/0733464813515089 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.