This paper is in the following e-collection/theme issue:

Public (e)Health, Digital Epidemiology and Public Health Informatics (636) Consumer & Patient Education and Shared-Decision Making (636) eHealth Literacy / Digital Literacy (351) Information Retrieval (271) Prevention and Health Promotion (879) Information Seeking, Information Needs (419)

Published on in Vol 17, No 11 (2015): November

The Association Between Online Health Information–Seeking Behaviors and Health Behaviors Among Hispanics in New York City: A Community-Based Cross-Sectional Study

The Association Between Online Health Information–Seeking Behaviors and Health Behaviors Among Hispanics in New York City: A Community-Based Cross-Sectional Study

The Association Between Online Health Information–Seeking Behaviors and Health Behaviors Among Hispanics in New York City: A Community-Based Cross-Sectional Study

Authors of this article:

Young Ji Lee1 Author Orcid Image ;   Bernadette Boden-Albala2 Author Orcid Image ;   Haomiao Jia3 Author Orcid Image ;   Adam Wilcox4 Author Orcid Image ;   Suzanne Bakken3, 5 Author Orcid Image
Article Authors Cited by (45) Tweetations (19) Metrics

    Original Paper

    1Department of Health and Community Systems, School of Nursing, University of Pittsburgh, Pittsburgh, PA, United States

    2Global Institute of Public Health, New York University, New York, NY, United States

    3School of Nursing, Columbia University, New York, NY, United States

    4Intermountain Healthcare, Murray, UT, United States

    5Department of Biomedical Informatics, Columbia University, New York, NY, United States

    Corresponding Author:

    Young Ji Lee, RN, MS, PhD

    Department of Health and Community Systems

    School of Nursing

    University of Pittsburgh

    3500 Victoria Street

    Pittsburgh, PA, 15261

    United States

    Phone: 1 412 624 7886

    Fax:1 412 383 7293

    Email: leeyoung@pitt.edu


    Background: Hispanics are the fastest-growing minority group in the United States and they suffer from a disproportionate burden of chronic diseases. Studies have shown that online health information has the potential to affect health behaviors and influence management of chronic disease for a significant proportion of the population, but little research has focused on Hispanics.

    Objective: The specific aim of this descriptive, cross-sectional study was to examine the association between online health information–seeking behaviors and health behaviors (physical activity, fruit and vegetable consumption, alcohol use, and hypertension medication adherence) among Hispanics.

    Methods: Data were collected from a convenience sample (N=2680) of Hispanics living in northern Manhattan by bilingual community health workers in a face-to-face interview and analyzed using linear and ordinal logistic regression. Variable selection and statistical analyses were guided by the Integrative Model of eHealth Use.

    Results: Only 7.38% (198/2680) of the sample reported online health information–seeking behaviors. Levels of moderate physical activity and fruit, vegetable, and alcohol consumption were low. Among individuals taking hypertension medication (n=825), adherence was reported as high by approximately one-third (30.9%, 255/825) of the sample. Controlling for demographic, situational, and literacy variables, online health information–seeking behaviors were significantly associated with fruit (β=0.35, 95% CI 0.08-0.62, P=.01) and vegetable (β=0.36, 95% CI 0.06-0.65, P=.02) consumption and physical activity (β=3.73, 95% CI 1.99-5.46, P<.001), but not alcohol consumption or hypertension medication adherence. In the regression models, literacy factors, which were used as control variables, were associated with 3 health behaviors: social networking site membership (used to measure one dimension of computer literacy) was associated with fruit consumption (β=0.23, 95% CI 0.05-0.42, P=.02), health literacy was associated with alcohol consumption (β=0.44, 95% CI 0.24-0.63, P<.001), and hypertension medication adherence (β=–0.32, 95% CI –0.62 to –0.03, P=.03). Models explained only a small amount of the variance in health behaviors.

    Conclusions: Given the promising, although modest, associations between online health information–seeking behaviors and some health behaviors, efforts are needed to improve Hispanics’ ability to access and understand health information and to enhance the availability of online health information that is suitable in terms of language, readability level, and cultural relevance.

    J Med Internet Res 2015;17(11):e261

    doi:10.2196/jmir.4368

    Keywords

    Internet; information-seeking behavior; health behavior; consumer health information; Hispanic Americans 


    Introduction

    Hispanics are the fastest-growing minority group in the United States and they suffer from a disproportionate burden of chronic diseases compared with non-Hispanics groups [1,2]. For instance, Hispanics have similar or lower levels of hypertension compared to non-Hispanic whites; however, their high blood pressure is less likely to be treated or controlled [3,4]. To manage illnesses and associated health problems, it is especially important for people with chronic diseases to engage in health-promoting behaviors [5-7]. However, studies have shown that there are disparities in health behaviors between Hispanics and non-Hispanics; Hispanics, as a group, spend less time on healthy behaviors than other ethnic groups [8-10]. For example, Hispanics report less physical activity time than non-Hispanics and have high rates of diseases related to sedentary lifestyle [9]. Also, compared with other ethnic groups, the rise in nutrition-related chronic diseases among Hispanics reflects an unequal burden of health issues [10].

    As the number of Internet users has grown, the Internet has become an important resource for chronic disease management [11]. Studies have shown that online health information has the potential to affect health behaviors and health outcomes for a significant proportion of the population [12,13]. Dutta-Bergman [14] demonstrated that online health information seekers reported higher levels of health behaviors, including physical activity, diet, alcohol consumption, medication, and smoking. In addition, online health information enables people to manage chronic diseases more effectively and improve their quality of life [15-17]. Because patient participation in health care decision making across the illness continuum is increasing, online health information–seeking behaviors are an important strategy for successful health interventions that inform decision making [18-20].

    Hispanics’ interest in online health information is increasing. In 2011, the Pew Internet & American Life Project reported that 45% of Hispanic Internet users had sought health information from the Internet; this increased to 66% in 2013 [21,22]. Compared with other ethnic groups, Hispanics are more likely to seek health information from the Internet or other social media rather than from a physician because they often lack access to a regular health care provider [23,24]. However, current studies have been limited to information seeking and its influence on health behaviors among the general public. Given the growth of Hispanic online health information seekers [25,26], examining the relationships between online health information–seeking behaviors and health behaviors in Hispanics is warranted.

    Our previous study focused on the correlates of online health information–seeking behaviors in Hispanics [27]. This study builds on that research by examining the association between online health information–seeking behaviors and 5 health behaviors (physical activity, fruit and vegetable consumption, alcohol use, and hypertension medication adherence) of relevance to Hispanics [2,14,28,29].


    Study Design and Hypotheses

    Data were collected during an in-depth community survey as part of the Washington Heights Inwood Informatics Infrastructure for Comparative Effectiveness Research (WICER) project. Five hypotheses were tested. Controlling for situational, sociodemographic, and literacy factors, online health information–seeking behaviors would be (1) positively associated with fruit consumption, (2) positively associated with vegetable consumption, (3) positively associated with physical activity, (4) negatively associated with alcohol consumption, and (5) positively associated with hypertension medication adherence. In addition, the associations between situational, sociodemographic, and literacy factors and fruit consumption, vegetable consumption, physical activity, alcohol consumption, and hypertension medication adherence were examined.

    Setting

    The survey catchment area included 5 ZIP codes (10031, 10032, 10033, 10034, and 10040) comprising the Washington Heights Inwood community of northern Manhattan, which has been designated a medically underserved area by the Centers for Medicare and Medicaid Services because it meets the criteria related to the level of poverty, elderly, and infant mortality, and the ratio of primary care providers in the community [30,31]. Currently, 71% of Washington Heights and Inwood area residents are Hispanic and 51% of residents are immigrants [32].

    Recruitment

    After approval by the Institutional Review Board of Columbia University Medical Center, recruitment of eligible participants was initiated using multiple methods. Participants were recruited from households or other places in the community (eg, schools, grocery stores, hair salons). The household sample started as a probability sample that evolved, as planned, to snowball sampling based on participants’ social networks and convenience sampling. For the probability sample, we randomized 68,000 dwelling units (ie, households) in the 5 ZIP codes and selected a weighted sample from each of the 8 New York City Department of Health and Mental Hygiene districts in the community. The sample size of each health district was weighted by the distribution of the population. Once we enrolled a household member and finished the survey, we asked participants if they would introduce the research team to members of their social network to ascertain interest in completing the survey (ie, snowball sampling). Other participants were recruited through direct contact with community health workers in community settings.

    Sample

    The sample consisted of those who completed the WICER community survey between April 2012 and November 2012. Participants who were 18 years or older, English or Spanish speaking, and Hispanic were eligible to participate in this study.

    Data Collection

    Data were collected via in-person interviews by bilingual community health workers in households or places in the community (eg, schools, grocery stores). For consistency of data collection across interviewers, community health workers received a combination of didactic and practical training in responsible conduct of research and specific study procedures including informed consent. Their performance was monitored by the project coordinator on a weekly basis. Before conducting the survey interview, bilingual study personnel obtained informed consent from the participant in their language of choice (English or Spanish). The process required approximately 45 to 60 minutes for completion and participants received US $25 compensation in the form of subway cards, grocery coupons, or movie tickets.

    Study Variables

    After WICER survey data collection was completed, Bodie and Dutta’s Integrative Model of eHealth Use [33] was applied to inform selection of correlates and related health outcomes for online health information–seeking behaviors from among the WICER variables and to guide the data analysis. This model includes demographic information, situational factors, health literacy, computer literacy, online health information–seeking behaviors, and health behavior variables. Table 1 summarizes the definitions of model concepts and operationalization of the concepts through WICER study variables and measures. As described in more detail in study measures, only selected dimensions of health literacy and computer literacy were measured in the study (ie, the available WICER variables were more narrow than the broad concepts in the model).

    Table 1. Conceptualization and measurement of study variables.
    Concept/DefinitionWICER Variable/Measure
    Situational factors: the specific health situations faced by a patient and his or her subsequent consumer health information needs [34]Hypertension: “Have you ever been told by a doctor, nurse, or other health professional that you had hypertension also called high blood pressure or pressure?” [35]

    		General health status: “Would you say that in general your health is...?” [36]

    		Serious health problems: “Have you experienced any serious personal health problems that have lasted for at least 6 months?” [37]
    Demographic informationAge, gender, employment, immigrant status, marital status, educational level, insurance
    Health literacy: the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions [38]Health literacy: Chew’s 1-item health literacy screening question “How often do you have problems learning about your medical condition because of difficulty understanding written information?” [39]
    Computer literacy: computer skills and ability to use technology to improve learning, productivity, and performance [40]Social networking site membership: “Do you belong to any social networking sites like Facebook, Myspace, or Twitter?”
    Online health information–seeking behaviors: the interaction of an individual with or through an electronic device or communication technology to access or transmit health information or to receive guidance and support on a health-related issue [41]Online health information–seeking behaviors: “In the past 12 months, have you (1) participated in an online support group for people with similar health or medical issues, (2) used email or the Internet to communicate with a doctor or doctor’s office, or (3) used the Internet to look up health or medical information?” [42]
    Health behaviorsPhysical activity, fruit consumption, vegetable consumption

    		Alcohol use: New York City Health and Nutrition Examination Survey (NYCHANES) [43]

    		Medication adherence: Morisky 8-Item Medication Adherence Scale (MMAS-8) [44]

    Bodie and Dutta [33] suggest that disparities in social structures, such as sociodemographic factors, lead to individual-level differences in online health information–seeking ability. This model also incorporates health literacy and computer literacy as predictors of online health information–seeking behaviors and health behaviors. In addition, it proposes that differences in health literacy are related to sociodemographic factors, such as age, race, birthplace, education level, and income, which also influence online health information–seeking behaviors [25,45]. Ultimately, this difference in online health information–seeking behaviors causes disparities in lifestyle that are related to health outcomes and continue to contribute to health care disparities [33].

    Demographic and Situational Factors

    As summarized in Table 1, situational and demographic data were obtained during the interview. For situational variables, hypertension was measured by the question “Have you ever been told by a doctor, nurse, or other health professional that you had hypertension also called high blood pressure or pressure?” [35]. The presence of serious health problems was measured by the question “Have you experienced any serious personal health problems that have lasted for at least 6 months?” from the Chronic Burden Scale [37]. Self-reported general health status was recorded in 5 categories: excellent, very good, good, fair, and poor. General health status was measured on a 5-point Likert scale (1=excellent, 5=poor) from the Short Form-8 Health Survey (SF-8) [36]. Demographic data included age, gender, employment, marital status, immigrant status, educational level, and type of insurance.

    Health Literacy

    Although the Institute of Medicine definition of health literacy is broad and encompasses accessing and processing as well as understanding and applying information, [38] we selected understanding as the focus of our health literacy measurement. To assess health literacy, we used a one-item health literacy screening measure: “How often do you have problems learning about your medical condition because of difficulty understanding written information?” Response options were always, often, sometimes, occasionally, and never. Criterion-related validity with Short Test of Functional Health Literacy in Adults (S-TOFHLA) was established by the area under the receiver operating characteristic curve (0.76) in a study of Veterans Affairs clinic patients [38]. Evidence for the reliability and validity of the 3 one-item scales from Chew et al [39] have been demonstrated in other populations, including females, African Americans [46], Turkish speakers [47], and Spanish speakers [48], This one-item scale was selected for the analysis in this paper instead of other WICER health literacy variables (eg, Newest Vital Sign which includes numeracy) because of its relevance to online health information–seeking behaviors. Moreover, Sarkar and colleagues [48] revalidated this single health literacy item among English and Spanish speakers and found similar test characteristics to previous studies; thus, it was appropriate for our primarily Spanish-speaking Hispanic participants. Per the method described by Chew and colleagues [39], the responses of always, often, and sometimes were coded as inadequate literacy; occasionally and never were coded as adequate literacy.

    Social Networking Site Membership

    There was no direct question regarding computer literacy in the WICER survey. The US Department of Education defines computer literacy as “computer skills and ability to use technology to improve learning, productivity, and performance” [40]. However, the definition of computer literacy changes as technology evolves [49] and several studies have included “using networked communication” as a component of computer literacy [50-52]. Use of social networking sites (SNS), whether by mobile device or desktop computer, requires a certain level of computer literacy as a foundation [53]. Therefore, based on the variables available in the WICER survey, one dichotomous question related to SNS membership was used to measure one dimension of computer literacy: “Do you belong to any social networking sites like Facebook, Myspace, or Twitter?” WICER survey variables specific to online health information–seeking behaviors also imply a level of computer literacy, but based on the Bodie and Dutta [33] model, were used to measure online health information–seeking behaviors.

    Online Health Information–Seeking Behaviors

    Robinson et al [41] defined interactive health communication as “the interaction of an individual—consumer, patient, caregiver, or professional—with or through an electronic device or communication technology to access or transmit health information or to receive guidance and support on a health-related issue.” Based on the definition, this study considered participation in an online support group, email communication with physicians, and using the Internet to look up health or medical information as online health information–seeking behaviors in this study. An affirmative response to at least 1 of 3 questions was coded as “yes” on online health information–seeking behaviors (no=0, yes=1).

    Health Behaviors

    Physical activity, fruit and vegetable consumption, and alcohol consumption were measured using items from New York City Health and Nutrition Examination Survey (NYCHANES) [43]. Moderate physical activity was measured by the question “Over the past 30 days, did you do moderate activities for at least 10 minutes that caused only light sweating or a slight to moderate increase in breathing or heart rate?” Answers were standardized into weekly rates. Similarly, fruit, vegetable, and dark vegetable consumption were standardized as number of servings per day. Alcohol use was first measured by the dichotomous question “Have you ever had alcoholic beverages such as beer, wine, champagne, or liquor at least once per month for 6 months or more?” If yes, the respondents were asked to select from among 9 frequencies of drinking. Responses were clustered into 3 categories: 1, 1-19, or ≥20 drinks per month [54]. Those who answered “no” to the initial question were categorized as <1.

    Hypertension medication adherence was measured by the Morisky 8-Item Medication Adherence Scale (MMAS-8) [44] with respondents explicitly instructed to limit the type of medication to those taken for hypertension. Three questions were dichotomous (no=0, yes=1) and 5 questions had 4 response options: (1) never or rarely, (2) once in a while, (3) sometimes, and (4) often. The former 2 were coded as “no” and the latter 2 as “yes.” One dichotomous item, “Did you take your high blood pressure medication yesterday?” required reverse coding. The responses were summed and adherence was defined as high (0), medium (1-2), or low (>2) [44]. Internal consistency reliability was good (α=.75) for the study sample.

    Statistical Analysis

    Data were analyzed using SPSS version 20.0 software. Univariate analyses were used to examine the frequency and distribution of study variables, calculating mean and standard deviation, range, frequency, and percentage as appropriate. The statistical analyses were guided by Bodie and Dutta’s Integrative Model of eHealth Use [33]. First, the relationships between demographics, situational and literacy (health literacy, SNS membership) variables, and online health information–seeking behaviors were examined using chi-square tests, t tests, and binary logistic regression with online health information–seeking behavior as the dependent variable. Although we previously conducted analyses of the variables associated with online health information–seeking behaviors in the larger WICER sample [27], analysis in this smaller sample was required to determine which variables should be controlled for in the hypothesis testing. Second, to test the study hypotheses, variables found to be significant in online health information–seeking behavior analyses were controlled for in regression models that examined the influence of online health information–seeking behaviors on health behaviors. Linear regression was used for continuous health behavior variables (physical activity and fruit and vegetable consumption). Ordinal regression models were used for alcohol use and hypertension medication adherence because those were coded as ordered categorical data. For each health behavior regression, 3 models were run: (1) demographic and situational variables only, (2) demographic and situational variables plus literacy variables, and (3) demographic, situational, and literacy variables plus online health information–seeking behaviors. The level of significance for testing of each model was set to an alpha of .05.


    Characteristics of Respondents

    Sample Characteristics

    The characteristics of the sample (N=2680) are summarized in Table 2. The mean age was 50.0 years (SD 17.1, range 18-100). Most were female (71.60%, 1919/2680) and immigrants (87.65%, 2349/2680). More than half of the respondents were unemployed (63.17%, 1693/2680), not married (64.33%, 1724/2680), and approximately half of the respondents had less than high school education (49.81%, 1335/2680). In all, 75.82% (2032/2680) of the participants were Medicare or Medicaid beneficiaries. Most respondents reported that their general health status as at least “good” (74.40%, 1994/2680) and that they did not have serious health problems (92.20%, 2471/2680). A total of 36.79% (986/2680) of respondents answered that they had been diagnosed with hypertension. More than the half of respondents (57.95%, 1553/2680) had inadequate health literacy and only 22.20% (595/2680) of respondents answered that they were a SNS member. Regarding online health information–seeking behaviors, only 7.38% (198/2680) of respondents reported that they had sought health information through the Internet. This was 29.0% (198/682) of the 682 respondents who reported using the Internet.

    Health Behaviors

    Respondents reported consuming fruit (mean 0.8, SD 1.5) and vegetables (mean 0.8, SD 1.7) less than once per day. The mean frequency of moderate physical activity was 1.6 times per week (SD 11.4). More than half of respondents (69.07%, 185/2680) reported consuming less than one alcoholic drink per month and a small percentage (2.0%, 54/2680) answered that they consumed more than 20 alcoholic drinks per month. Among individuals taking hypertension medication (n=825), adherence was reported as high (30.9%, 255/825), medium (26.3%, 217/825), and low (42.8%, 353/825).

    Relationships Between Demographic, Situational, and Literacy Variables and Online Health Information–Seeking Behaviors

    In bivariate analyses related to demographic and situational variables and respondents’ online health information–seeking behaviors, there were statistically significant differences in age (P=.001), education (P<.001), employment status (P<.001), hypertension (P<.001), insurance (P=.005), immigrant status (P<.001), general health status (P<.001), health literacy (P<.001), and SNS membership (P<.001). Other demographic and situational variables were also significant in the binomial logistic regressions: older age (OR 1.68, 95% CI 1.29-2.20, P<.001), a high level of education (OR 3.07, 95% CI 1.99-4.80, P<.001), being born in the United States (OR 1.68, 95% CI 1.10-2.56, P=.02), worse health status (OR 0.39, 95% CI 0.27-0.57, P<.001), and lack of hypertension (OR 0.64, 95% CI 0.42-0.99, P=.047) were associated with participants’ online health information–seeking behaviors. Controlling for these variables, respondents with adequate health literacy (OR 2.13, 95% CI 1.52-2.99, P<.001) and SNS members (OR 4.21, 95% CI 2.86-6.19, P<.001) had increased odds of online health information–seeking behaviors.

    Hypotheses Testing: Relationships Between Online Health Information–Seeking Behaviors and Health Behaviors Controlling for Demographic, Situational, and Literacy Factors

    Models 1 and 2 in Tables 3-7 are displayed to reflect the relationships between the final regressions and the eHealth use framework that guided the analysis. Model 1 shows the results of the demographic and situational variables regressed on the health behaviors and Model 2 adds the literacy variables. In terms of hypothesis testing, controlling for demographic, situational, and literacy factors (as shown in Model 3 in Tables 3-7), online health information–seeking behaviors were positively associated with fruit consumption (β=0.35, 95% CI 0.08-0.62, P=.01; Table 3), vegetable consumption (β=0.36, 95% CI 0.06-0.65, P=.02; Table 4), and physical activity (β=3.73, 95% CI 1.99-5.46, P<.001; Table 5). Thus, hypotheses 1-3 were supported. Other variables significantly related with health behaviors in Model 3 were age (β=–0.12, 95% CI –0.23 to –0.01, P=.04), immigrant status (β=–0.33, 95% CI –0.56 to –0.10, P=.01), general health status (β=–0.24, 95% CI –0.41 to –0.08, P=.01), and SNS membership (β=0.23, 95% CI 0.05-0.42, P=.02) for fruit consumption; immigrant status for vegetable consumption (β=–0.33, 95% CI –0.59 to –0.08, P=.01); and age (β=–0.87, 95% CI –1.60 to –0.13, P=.02), general health status (β=–1.92, 95% CI –3.00 to –0.83, P=.001), and education level (β=1.28, 95% CI 0.29-2.27, P=.01) for physical activity. However, each model explained only a small amount of the variance in each health behavior.

    Table 2. Descriptive characteristics of Hispanic participants (N=2680).
    VariablesRespondents
    Demographic factors

    		Gender, n (%)


    		Men747 (27.87)


    		Women1919 (71.60)

    		Employment status, n (%)


    		Employed987 (36.83)


    		Unemployed1693 (63.17)

    		Marital status, n (%)


    		Married/living as935 (34.89)


    		Otherwise1724 (64.33)

    		Education, n (%)


    		<High school graduate1335 (49.81)


    		≥High school graduate1298 (48.43)

    		Birthplace, n (%)


    		Born in the United States323 (12.05)


    		Born in the other countries2349 (87.65)

    		Insurance, n (%)


    		Medicare/Medicaid2032 (75.82)


    		Others (VA, private, etc)392 (14.63)


    		None356 (13.28)

    		Age (years), mean (SD)50.0 (17.1)
    Situational factors

    		General health status, n (%)


    		<Good612 (22.84)


    		≥Good1994 (74.40)

    		Hypertension, n (%)


    		Yes986 (36.79)


    		No1662 (62.01)

    		Serious personal health problems, n (%)


    		Yes187 (6.98)


    		No2471 (92.20)
    Literacy factors

    		Health literacy, n (%)


    		Adequate literacy1009 (37.65)


    		Inadequate literacy1553 (57.95)

    		SNS membership, n (%)


    		Yes595 (22.20)


    		No1750 (65.30)

    		Online health information–seeking behaviors, n (%)


    		Yes198 (7.39)


    		No2471 (92.20)
    Health behaviors

    		Physical activity (times/week), mean (SD)1.6 (11.4)

    		Fruit consumption (times/day), mean (SD)0.8 (1.5)

    		Vegetable consumption (times/day), mean (SD)0.8 (1.7)

    		Alcohol consumption, n (%)


    		<1 per month1851 (69.07)


    		1-19 per month660 (24.63)


    		≥20 per month54 (2.01)

    		Medication adherence,an (%)


    		High adherence255 (30.9)


    		Medium adherence217 (26.3)


    		Low adherence353 (42.8)

    a Sample size for medication adherence is n=825.

    Table 3. Linear regression: association between online health information–seeking behaviors and fruit consumption.
    VariablesModel 1aModel 2bModel 3b

    		β (SE)95% CIPβ (SE)95% CIPβ (SE)95% CIP
    Demographic and situational factors









    		Immigrant statusc–0.24 (0.12)–0.47, –0.01.04–0.31 (0.12)–0.54, –0.08.01–0.33 (0.12)–0.56, –0.09.01

    		Age–0.14 (0.06)–0.26, –0.03.01–0.11 (0.06)–0.22, 0.01.07–0.12 (0.06)–0.24, –0.01.04

    		General health status–0.27 (0.09)–0.44, –0.10.002–0.27 (0.09)–0.44, –0.10.002–0.24 (0.09)–0.41, –0.08.01

    		Hypertensiond–0.01 (0.08)–0.17, 0.15.920.02 (0.08)–0.14, 0.18.810.03 (0.08)–0.13, 0.19.70

    		Education level0.13 (0.08)–0.03, 0.28.100.08 (0.08)–0.08, 0.23.320.05 (0.08)–0.10, 0.21.50
    Literacy factors









    		Health literacy


    		–0.03 (0.07)–0.15, 0.14.97–0.02 (0.07)–0.17, 0.12.76

    		SNS membershipd


    		0.28 (0.09)0.10, 0.47.010.23 (0.10)0.05, 0.42.02
    Online health information–seeking behaviorsd





    		0.35 (0.14)0.08, 0.62.01

    a Adjusted R2=.01, P=.001.

    b Adjusted R2=.01, P<.001.

    c Birthplace was coded as 0=foreign born or 1=US born.

    d Coded as 0=no or 1=yes.

    Table 4. Linear regression: association between online health information–seeking behaviors and vegetable consumption.
    VariablesModel 1aModel 2bModel 3c

    		β (SE)95% CIPβ (SE)95% CIPβ (SE)95% CIP
    Demographic and situational factors









    		Immigrant statusc–0.28 (0.13)–0.53, –0.02.03–0.31 (0.13)–0.57, –0.06.02–0.33 (0.13)–0.59, –0.08.01

    		Age–0.10 (0.06)–0.22, 0.02.11–0.07 (0.06)–0.19, 0.06.28–0.09 (0.06)–0.21, 0.04.19

    		General health status–0.09 (0.09)–0.28, 0.09.31–0.11 (0.09)–0.29, 0.08.26–0.08 (0.09)–0.26, 0.11.41

    		Hypertensiond–0.04 (0.09)–0.21, 0.14.68–0.02 (0.09)–0.19, 0.16.85–0.01 (0.09)–0.18, 0.17.95

    		Education level0.11 (0.08)–0.06, 0.27.210.07 (0.09)–0.10, 0.24.400.05 (0.09)–0.12, 0.22.60
    Literacy factors









    		Health literacy


    		–0.12 (0.08)–0.28, 0.04.14–0.14 (0.08)–0.30, 0.02.08

    		SNS membershipd


    		0.20 (0.10)–0.001, 0.40.050.15 (0.11)–0.06, 0.36.15
    Online health information–seeking behaviorsd





    		0.36 (0.15)0.06, 0.65.02

    a Adjusted R2=.002, P=.12.

    b Adjusted R2=.004, P=04.

    c Adjusted R2=.006, P=.009.

    c Immigrant status was coded as 0=foreign born or 1=US born.

    d Coded as 0=no or 1=yes.

    Table 5. Linear regression: association between online health information–seeking behaviors and physical activity.
    VariablesModel 1aModel 2aModel 3b

    		β (SE)95% CIPβ (SE)95% CIPβ (SE)95% CIP
    Demographic and situational factors









    		Immigrant statusc–0.90 (0.75)–2.36, 0.57.23–1.19 (0.76)–2.68, 0.30.12–1.40 (0.76)–2.89, 0.08.06

    		Age–0.83 (0.37)–1.54, –0.11.02–0.70 (0.38)–1.43, 0.04.002–0.87 (0.38)–1.60, –0.13.02

    		General health status–2.30 (0.55)–3.37, –1.21<.001–2.22 (0.56)–3.23, –1.12<.001–1.92 (0.55)–3.00, –0.83.001

    		Hypertensiond0.47 (0.52)–0.56, 1.45.370.58 (0.53)–0.45, 1.61.270.71 (0.53)–0.32, 1.74.18

    		Education level1.75 (0.50)0.78, 2.72<.0011.55 (0.50)0.56, 2.54.0021.28 (0.51)0.29, 2.27.01
    Literacy factors









    		Health literacy


    		0.99 (0.47) a0.07, 1.91.040.77 (0.47)–0.1, 1.70.10

    		SNS membershipd


    		1.16 (0.60)–0.02, 2.34.050.67 (0.61)–0.53, 1.86.28
    Online health information–seeking behaviorsd





    		3.73 (0.88)1.99, 5.46<.001

    a Adjusted R2=.01, P<.001.

    b Adjusted R2=.02, P<.001.

    c Immigrant status was coded as 0=foreign born or 1=US born.

    d Coded as 0=no or 1=yes.

    The hypotheses related to the association between online health information–seeking behaviors and alcohol consumption (Table 6) and hypertension medication adherence (Table 7) were not supported. For other variables in Model 3, health literacy (β=0.44, 95% CI 0.24-0.63, P<.001) was significantly associated with alcohol consumption and the final model explained 2.0% of the variance in alcohol consumption (R2=.02, P<.001). Age (β=–0.27, 95% CI –0.51 to –0.02, P=.04), education level (β=–0.58, 95% CI –0.90 to –0.25, P<.001), and health literacy (β=–0.32, 95% CI –0.62 to –0.03, P=.03) were significantly associated with hypertension medication adherence and Model 3 explained 5.0% of the variance (R2=.05, P<.001).

    Table 6. Ordinal logistic regression: association between online health information–seeking behaviors and alcohol consumption.
    VariablesModel 1aModel 2bModel 3b

    		β (SE)95% CIPβ (SE)95% CIPβ (SE)95% CIP
    Demographic and situational factors









    		Immigrant statusc0.24 (0.14)–0.04, 0.52.100.25 (0.16)–0.05, 0.55.110.23 (0.16)–0.07, 0.54.13

    		Age0.07 (0.07)–0.07, 0.21.330.13 (0.08)–0.03, 0.28.100.11 (0.08)–0.04, 0.27.16

    		General health status0.24 (0.11)0.02, 0.46.030.20 (0.12)–0.03, 0.43.090.21 (0.12)0.02, 0.45.08

    		Hypertensiond–0.03 (0.10)–0.24, 0.17.75–0.03 (0.11)–0.25, 0.19.78–0.02 (0.11)–0.24, 0.20.87

    		Education level0.08 (0.10)–0.11, 0.27.390.05 (0.11)–0.16, 0.26.610.04 (0.11)–0.17, 0.26.68
    Literacy factors









    		Health literacy


    		0.46 (0.10)0.26, 0.65<.0010.44 (0.10) c0.24, 0.63.<001

    		SNS membershipd


    		0.17 (0.13)–0.08, 0.41.190.13 (0.13) c–0.12, 0.39.29
    Online health information–seeking behaviorsd





    		–0.17 (0.18)–0.52, 0.19.35

    a Adjusted R2=.01, P=.07 .

    b Adjusted R2=.02, P<.001.

    c Immigrant status was coded as 0=foreign born or 1=US born.

    d Coded as 0=no or 1=yes.

    Table 7. Ordinal logistic regression: association between online health information–seeking behaviors and hypertension medication adherence (n=825).
    VariablesModel 1aModel 2bModel 3c

    		β (SE)95% CIPβ (SE)95% CIPβ (SE)95% CIP
    Demographic and situational factors









    		Immigrant statusd1.31 (0.53)0.27, 2.34.010.89 (0.57)–0.23, 2.00.150.90 (0.57)–0.22, 2.02.12

    		Age–0.29 (0.12)–0.51,–0.06.01–0.28 (0.13)–0.52, –0.03.03–0.27 (0.13)–0.51, –0.02.04

    		General health statuse0.33 (0.14)0.05, 0.61.020.22 (0.15)–0.08, 0.52.150.24 (0.16)–0.01, 0.54.13

    		Education level–0.40 (0.15)–0.69,–0.11.01–0.55 (0.16)–0.87, –0.23.001–0.58 (0.17) c–0.90, –0.25<.001
    Literacy factors









    		Health literacy


    		–0.32 (0.15)–0.61,–0.03.03–0.32 (0.15)–0.62, –0.03.03

    		SNS membershipe


    		0.44 (0.29)–0.14, –1.00.130.41 (0.29)–0.17, 0.98.17
    Online health information–seeking behaviorse





    		–0.64 (0.40)–1.42, 0.14.11

    a Adjusted R2=.03, P<.001.

    b Adjusted R2=.04, P<.001.

    c Adjusted R2=.05, P=.

    d Immigrant status was coded as 0=foreign born or 1=US born.

    e Coded as 0=no or 1=yes.


    Principal Results

    Only 7.38% (198/2680) of respondents reported that they had sought health information through the Internet; this was 29.0% (198/682) of the 682 respondents who indicated that they used the Internet. This is significantly different than the 66% of Hispanic Internet users who reported online health information–seeking behaviors according to the Pew Internet & American Life Project in 2013 [21]. The difference may be a reflection of the low socioeconomic status of the Washington Heights and Inwood area in this study that may have limited participants’ ability to pay for Internet access. Alternatively, we did not specifically ask about accessing the Internet via cellular telephone and respondents may have failed to consider this mode of access in their response. Thus, the actual rate of online health information–seeking behaviors in our sample may be higher than we captured in the WICER survey.

    Guided by Bodie and Dutta’s Integrative Model of eHealth Use [33], demographic, situational, and literacy factors were examined to determine which variables were significantly correlated with online health information–seeking behaviors as a preliminary step to determine which variables to control for in hypothesis testing of the relationship between online health information–seeking behaviors and health behaviors. Significant demographic variables associated with online health information–seeking behaviors were older age, higher level of education, and being born in the United States. Education level and immigrant status were consistent with previous studies [55-58]; however, the relationship of age to online health information–seeking behaviors contrasted with existing studies [55-57]. This may be explained by the increasing interest in online health information among older adults. More than half of our sample was older than age 40 years (1850/2680, 69.03%), and 19.70% (528/2680) were older than age 65 years. Consistent with our prior analysis using the larger WICER sample [27], worse health status and lack of hypertension were associated with participants’ online health information–seeking behaviors.

    Previous research has provided evidence that online health information seekers tend to improve their health behaviors [59-61]. Controlling for demographic, situational, and literacy factors, this study tested 5 hypotheses about the association between Hispanics’ online health information–seeking behaviors and health behaviors: fruit consumption, vegetable consumption, physical activity, alcohol consumption, and hypertension medication adherence. Although only a small proportion of the sample reported online health information–seeking behaviors, online health information–seeking behaviors were positively associated with fruit and vegetable consumption and physical activity. Our study showed that online health information seekers were likely to consume more fruits and vegetables and report more physical activity than nonseekers. However, the average fruit and vegetable consumption and physical activity levels were significantly less than the Centers for Disease Control and Prevention guidelines [62]. Moreover, the models explained only a small portion of the variance. Consequently, these findings, although promising, must be interpreted cautiously.

    In contrast, online health information–seeking behaviors were not significantly associated with alcohol consumption or adherence to hypertension medication, which is inconsistent with previous studies [63-66]. There are several possible reasons for this. First, other studies did not include a large number of immigrant Hispanics and findings may vary according to these demographic characteristics. Second, there may be lack of online content on medication adherence and alcohol consumption targeted for Hispanics, who were the focus of our study, and that could limit the motivation to seek such information online. Third, a study by Shabab et al [66] suggested that online health information may not be equally effective for all types of health behaviors, which may shed light on why our findings differed for various health behaviors.

    In addition to the hypothesis testing, the study revealed relationships between the demographic, situational, and literacy factors, which served as control variables in the regressions related to the 5 health behaviors. There were significant relationships between health behaviors and the demographic characteristics of age (fruit and vegetable consumption, medication adherence), education level (physical activity, hypertension medication adherence), and immigrant status (fruit and vegetable consumption). In terms of immigrant status, consistent with previous research [67,68], foreign-born Hispanic participants were likely to eat more fruit and vegetables than US-born participants. Our results showed that younger people were likely to eat more fruit and have higher physical activity levels than older people, which contrasts with existing literature [69-71]. Our findings also showed that individuals with higher education were more likely to engage in physical activity and more likely to adhere to hypertension medication regimens. In regards to the latter, findings related to education level and age were inconsistent with existing literature [72,73] that showed that lower educational attainment and older age were positively associated with medication adherence. One reason for the differences may be that our study specifically focused on hypertension medication adherence rather than medication adherence in general.

    In terms of situational factors, poor health status was associated with more fruit consumption and physical activity. This is consistent within Shim’s [74] finding that people who perceived their health status as low were more likely to change their behavior after searching for health information online. Although a previous study showed having hypertension was significantly associated with health behaviors such as alcohol consumption and physical activities [75], our study did not show a significant association between hypertension and health behaviors.

    Additionally, literacy factors were associated with some health behaviors: SNS membership was associated with fruit consumption and health literacy was associated with alcohol consumption and medication adherence. Previous research has reported that study participants who frequently visit SNS were willing to communicate with others about health-related issues and exchange health information through those sites [76-78]. A large survey based on 23,000 people in the United States indicated that 41% of respondents had used any social media as a health information resource; 94% of those users chose Facebook to seek health-related information [77]. Consequently, one explanation for the association between SNS membership and health behaviors is that SNS members may use those sites to find or discuss health information related to healthy diet, such as eating fruit. SNS have the potential be used as a channel to convey health information to promote health behavior, especially tailored to SNS members through user profiles, forums, blogs, comments, search queries, and tags [79].

    Moreover, our study findings that participants with higher health literacy were more likely to consume less alcohol [80-82] and to adhere to hypertension medication were consistent with previous studies [83,84]. Differences in health literacy are often related to sociodemographic factors that also influence health behaviors [33,45]. Hispanics are an important target in efforts to improve health literacy due to their lower health literacy levels when compared to other ethnic groups [85]. Adequate health literacy is a prerequisite to directly influence health behaviors and to affect online health information–seeking behavior as a strategy for improving health behaviors [86]. In fact, although we conceptualized it separately in our study, effective health information–seeking behavior is one dimension of the Institute of Medicine definition of health literacy [38].

    The significant associations in our study between online health information–seeking behaviors and some health behaviors as well as between our measures of literacy factors (health, SNS membership) and fruit consumption, alcohol consumption, and hypertension suggest that strategies are needed in 2 areas to promote use of online health information by Hispanics. First, skills related to health literacy and online health information–seeking behaviors must be strengthened. Finding health information online challenges those with poor health literacy and limited skills related to online health information–seeking behaviors [87-89] and most existing search tools require high levels of computer literacy for optimal use [90,91]. Second, understandable online health information is necessary to fulfill the information needs of Hispanics so that online health information is perceived as useful [92-94]. More than half of our sample had difficulty understanding written information related to their medical problems. Moreover, online health information is rarely provided in Spanish [95] and Spanish online health information is often either at an advanced reading level or of poor quality [88]. Several authors suggest that the reading level for online information should be at the grade 7 or 8 level [93,94], but that may be too high for those with low health literacy. Additionally, although not explicitly examined in this study because all participants were Hispanic and predominantly from the Dominican Republic, experts recommend that the cultural beliefs and values of the designated population also should be reflected in the content of online health information to encourage ease of use and health behavior change [96,97]. Our study may provide a foundation to understand the characteristics of Hispanics living in the Washington Heights and Inwood community, which may support the dissemination of online information more strategically [11].

    Limitations

    This study has several potential limitations. First, the generalizability is limited due to the nonprobability sample and the fact that the Hispanics in our sample were primarily from the Dominican Republic; our sample does not reflect the heterogeneity of socioeconomic status, culture, and health of Hispanics in the United States [11,34,98,99]. Second, the ratio of Internet users who reported online health information–seeking behaviors may be artificially high; the number of Internet users is most likely underestimated because it was not directly asked in the WICER survey. Rather, it was calculated from the unique number of respondents who indicated either online health information–seeking behaviors or SNS use. Third, the data were cross-sectional rather than longitudinal, thus it is not possible to draw causal inferences between online health information–seeking behaviors and health behaviors. Fourth, because all measures were self-reported, recall bias and social desirability bias [100] may have influenced participants’ responses. Fifth, our measures of health literacy and computer literacy were narrower than the broader definitions in the conceptual model used for selection and analysis of variables. In particular, SNS membership alone in the absence of measurements of use may not accurately reflect overall level of computer literacy. Sixth, the survey was conducted by face-to-face interview by multiple community health workers. Although the interviewers were carefully trained and monitored, differences among interviewers may have affected the survey result [101]. However, face-to-face survey administration by bilingual community health workers was critical to the success of the WICER study given the documented underrepresentation of Hispanics in research and the low levels of health literacy in the study sample. Lastly, the questions related to online health information–seeking behaviors did not explicitly query the use of cellular phones and participants may not have considered this avenue of accessing the Internet in their response. This may be particularly important given that Hispanics are more likely to access the Internet from mobile phones than from desktop computers [21]. For example, the Pew Hispanic Center reported that Hispanics are more likely to access the Internet through mobile devices compared to non-Hispanics; 76% of Hispanics access the mobile Internet compared to 60% of non-Hispanics [102]. Because we only asked participants about the use of “Internet” there is the possibility that participants did not count the use of mobile Internet as the use of Internet. Therefore, this study may have underestimated online health information–seeking behaviors among the survey respondents.

    Conclusions

    To our knowledge, this is the first large-scale study to examine the relationship between online health information–seeking behaviors and health behaviors in the Hispanic population. Although conclusions from a single study should be interpreted carefully, the data in this study suggest potential avenues for informatics-based health interventions. Given the promising, although modest, associations between online health information–seeking behaviors and 3 health behaviors (fruit and vegetable consumption and physical activity) efforts are needed to improve Hispanics’ ability to access and understand health information and to enhance the availability of online health information that is suitable in terms of language, readability level, and culture.

    Acknowledgments

    This study was supported by the Agency for Healthcare Research and Quality (R01HS019853, R01HS022961, Bakken, PI) and was completed as part of the PhD thesis of Young Ji Lee at Columbia University School of Nursing.

    Conflicts of Interest

    None declared.

    1. Lopez-Quintero C, Berry EM, Neumark Y. Limited English proficiency is a barrier to receipt of advice about physical activity and diet among Hispanics with chronic diseases in the United States. J Am Diet Assoc 2009 Oct;109(10):1769-1774. [CrossRef] [Medline]
    2. Viruell-Fuentes EA, Ponce NA, Alegría M. Neighborhood context and hypertension outcomes among Latinos in Chicago. J Immigr Minor Health 2012 Dec;14(6):959-967. [CrossRef] [Medline]
    3. National Research Council (US) Panel on Hispanics in the United States. Hispanics and the Future of America. Washington, DC: National Academies Press; 2006.
    4. American Heart Association. Bridging the Gap: CVD Health Disparities. 2011.   URL: http://www.heart.org/idc/groups/heart-public/@wcm/@hcm/@ml/documents/downloadable/ucm_429240.pdf [accessed 2015-04-20] [WebCite Cache]
    5. McGinnis JM, Williams-Russo P, Knickman JR. The case for more active policy attention to health promotion. Health Aff (Millwood) 2002;21(2):78-93 [FREE Full text] [Medline]
    6. Manuvinakurike R, Velicer WF, Bickmore TW. Automated indexing of Internet stories for health behavior change: weight loss attitude pilot study. J Med Internet Res 2014;16(12):e285 [FREE Full text] [CrossRef] [Medline]
    7. Tucker CM, Wippold GM, Smith TM, Desmond FF, Arthur TM, Butler AM, et al. Association of health self-empowerment with health-promoting behaviors among chronically ill African American and non- Hispanic White adolescents. J Health Care Poor Underserved 2014 Nov;25(4):2019-2031. [CrossRef] [Medline]
    8. Perez-Escamilla R. Health care access among Latinos: implications for social and health care reforms. Journal of Hispanic Higher Education 2009 Dec 29;9(1):43-60. [CrossRef]
    9. Larsen BA, Pekmezi D, Marquez B, Benitez TJ, Marcus BH. Physical activity in Latinas: social and environmental influences. Womens Health (Lond Engl) 2013 Mar;9(2):201-210 [FREE Full text] [CrossRef] [Medline]
    10. Kerber C, Kessler L, Wallace S, Burns-Whitmore B. Cultural and dietary factors influencing traditional Latino meal patterns: findings from focus group discussion. Californian Journal of Health Promotion 2014;12(2):42-54 [FREE Full text]
    11. Cheong PH. Health communication resources for uninsured and insured Hispanics. Health Commun 2007;21(2):153-163. [CrossRef] [Medline]
    12. Chisolm DJ. Does online health information seeking act like a health behavior?: a test of the behavioral model. Telemed J E Health 2010 Mar;16(2):154-160. [CrossRef] [Medline]
    13. Gilmour JA. Reducing disparities in the access and use of Internet health information. a discussion paper. Int J Nurs Stud 2007 Sep;44(7):1270-1278. [CrossRef] [Medline]
    14. Dutta-Bergman MJ. Health attitudes, health cognitions, and health behaviors among Internet health information seekers: population-based survey. J Med Internet Res 2004 May 28;6(2):e15 [FREE Full text] [CrossRef] [Medline]
    15. Kalichman SC, Cherry C, Cain D, Pope H, Kalichman M, Eaton L, et al. Internet-based health information consumer skills intervention for people living with HIV/AIDS. J Consult Clin Psychol 2006 Jun;74(3):545-554. [CrossRef] [Medline]
    16. Morrison D, Wyke S, Agur K, Cameron EJ, Docking RI, Mackenzie AM, et al. Digital asthma self-management interventions: a systematic review. J Med Internet Res 2014;16(2):e51 [FREE Full text] [CrossRef] [Medline]
    17. Stellefson M, Chaney B, Barry AE, Chavarria E, Tennant B, Walsh-Childers K, et al. Web 2.0 chronic disease self-management for older adults: a systematic review. J Med Internet Res 2013 Feb;15(2):e35 [FREE Full text] [CrossRef] [Medline]
    18. Alpay L, Verhoef J, Toussaint P, Zwetsloot-Schonk B. What makes an "informed patient"? The impact of contextualization on the search for health information on the Internet. Stud Health Technol Inform 2006;124:913-919. [Medline]
    19. Anker AE, Reinhart AM, Feeley TH. Health information seeking: a review of measures and methods. Patient Educ Couns 2011 Mar;82(3):346-354. [CrossRef] [Medline]
    20. Flynn KE, Smith MA, Freese J. When do older adults turn to the internet for health information? Findings from the Wisconsin Longitudinal Study. J Gen Intern Med 2006 Dec;21(12):1295-1301 [FREE Full text] [CrossRef] [Medline]
    21. Fox S, Duggan M. Health Online 2013. Washington, DC: Pew Internet & American Life Project; 2013 Jan 15.   URL: http://www.pewinternet.org/2013/01/15/health-online-2013/ [WebCite Cache]
    22. Fox S. The Social Life of Health Information, 2011. Washington, DC: Pew Internet & American Life Project; 2011 May 12.   URL: http://www.pewinternet.org/files/old-media/Files/Reports/2011/PIP_Social_Life_of_Health_Info.pdf [accessed 2015-04-20] [WebCite Cache]
    23. Livingston G. The Latino Digital Divide: The Native Born Versus the Foreign Born. Washington, DC: Pew Hispanic Center; 2010 Jul 28.   URL: http://www.pewhispanic.org/files/reports/123.pdf [accessed 2014-04-28] [WebCite Cache]
    24. Livingston G, Minushkin S, Cohn C. Hispanics and Health Care in the United States: Access, Information and Knowledge. Washington, DC: Pew Hispanic Center; 2008 Aug.   URL: http://www.pewhispanic.org/files/reports/91.pdf [accessed 2015-02-23] [WebCite Cache]
    25. Spittaels H, De Bourdeaudhuij I, Brug J, Vandelanotte C. Effectiveness of an online computer-tailored physical activity intervention in a real-life setting. Health Educ Res 2007 Jun;22(3):385-396 [FREE Full text] [CrossRef] [Medline]
    26. Lewis N, Martinez LS, Freres DR, Schwartz JS, Armstrong K, Gray SW, et al. Seeking cancer-related information from media and family/friends increases fruit and vegetable consumption among cancer patients. Health Commun 2012;27(4):380-388 [FREE Full text] [CrossRef] [Medline]
    27. Lee YJ, Boden-Albala B, Larson E, Wilcox A, Bakken S. Online health information seeking behaviors of Hispanics in New York City: a community-based cross-sectional study. J Med Internet Res 2014;16(7):e176 [FREE Full text] [CrossRef] [Medline]
    28. Bailey GR, Barner JC, Weems JK, Leckbee G, Solis R, Montemayor D, et al. Assessing barriers to medication adherence in underserved patients with diabetes in Texas. Diabetes Educ 2012;38(2):271-279. [CrossRef] [Medline]
    29. Pellowski JA, Kalichman SC. Health behavior predictors of medication adherence among low health literacy people living with HIV/AIDS. J Health Psychol 2015 Feb 20:epub. [CrossRef] [Medline]
    30. Pati S, Romero D, Chavkin W. Changes in use of health insurance and food assistance programs in medically underserved communities in the era of welfare reform: an urban study. Am J Public Health 2002 Sep;92(9):1441-1445. [Medline]
    31. Zach L, Dalrymple PW, Rogers ML, Williver-Farr H. Assessing internet access and use in a medically underserved population: implications for providing enhanced health information services. Health Info Libr J 2012 Mar;29(1):61-71. [CrossRef] [Medline]
    32. New York City Department of Health and Mental Hygiene. Community Health Profiles: Inwood and Washington Heights. 2006.   URL: http://www.nyc.gov/html/doh/downloads/pdf/data/2006chp-301.pdf [WebCite Cache]
    33. Bodie GD, Dutta MJ. Understanding health literacy for strategic health marketing: eHealth literacy, health disparities, and the digital divide. Health Mark Q 2008 Jan;25(1-2):175-203. [CrossRef] [Medline]
    34. McDonald SP, Collins JF, Johnson DW. Obesity is associated with worse peritoneal dialysis outcomes in the Australia and New Zealand patient populations. J Am Soc Nephrol 2003 Nov;14(11):2894-2901 [FREE Full text] [Medline]
    35. 2013 Behavioral Risk Factor Surveillance System Survey Questionnaire. 2012 Dec 28.   URL: http://www.cdc.gov/brfss/questionnaires/pdf-ques/2013%20BRFSS_English.pdf [accessed 2014-04-28] [WebCite Cache]
    36. Yen ST, Chen Z, Eastwood DB. Lifestyles, demographics, dietary behavior, and obesity: a switching regression analysis. Health Serv Res 2009 Aug;44(4):1345-1369 [FREE Full text] [CrossRef] [Medline]
    37. Mujahid MS, Diez Roux AV, Cooper RC, Shea S, Williams DR. Neighborhood stressors and race/ethnic differences in hypertension prevalence (the Multi-Ethnic Study of Atherosclerosis). Am J Hypertens 2011 Feb;24(2):187-193 [FREE Full text] [CrossRef] [Medline]
    38. Nielsen-Bohlman L, Panzer A, Kindig D. Health Literacy: A Prescription to End Confusion. Washington, DC: National Academies Press; 2004.
    39. Chew LD, Bradley KA, Boyko EJ. Brief questions to identify patients with inadequate health literacy. Fam Med 2004 Sep;36(8):588-594 [FREE Full text] [Medline]
    40. Bers MU. Beyond computer literacy: supporting youth's positive development through technology. New Dir Youth Dev 2010;2010(128):13-23. [CrossRef] [Medline]
    41. Robinson TN, Patrick K, Eng TR, Gustafson D. An evidence-based approach to interactive health communication: a challenge to medicine in the information age. Science Panel on Interactive Communication and Health. JAMA 1998 Oct 14;280(14):1264-1269. [Medline]
    42. National Cancer Institute. 2007. Health Information National Trends Survey: survey instruments   URL: http://hints.cancer.gov/instrument.aspx [accessed 2014-04-28] [WebCite Cache]
    43. The New York City Department of Health and Mental Hygiene. NYC HANES datasets and related documentation   URL: http://www.nyc.gov/html/doh/html/data/hanes-datasets.shtml [WebCite Cache]
    44. Morisky DE, Ang A, Krousel-Wood M, Ward HJ. Predictive validity of a medication adherence measure in an outpatient setting. J Clin Hypertens (Greenwich) 2008 May;10(5):348-354 [FREE Full text] [Medline]
    45. Rudd R, Kirsch I, Yamamoto K. Literacy and Health in America. Princeton, NJ: Educational Testing Service; 2004.   URL: http://files.eric.ed.gov/fulltext/ED486416.pdf [accessed 2015-04-20] [WebCite Cache]
    46. Schwartz KL, Bartoces M, Campbell-Voytal K, West P, Monsur J, Sartor A, et al. Estimating health literacy in family medicine clinics in metropolitan Detroit: a MetroNet study. J Am Board Fam Med 2013;26(5):566-570 [FREE Full text] [CrossRef] [Medline]
    47. Eyüboğlu E, Schulz PJ. Validation of Turkish health literacy measures. Health Promot Int 2015 Jan 13:epub. [CrossRef] [Medline]
    48. Sarkar U, Piette JD, Gonzales R, Lessler D, Chew LD, Reilly B, et al. Preferences for self-management support: findings from a survey of diabetes patients in safety-net health systems. Patient Educ Couns 2008 Jan;70(1):102-110 [FREE Full text] [CrossRef] [Medline]
    49. Lin T. A computer literacy scale for newly enrolled nursing college students: development and validation. J Nurs Res 2011 Dec;19(4):305-317. [CrossRef] [Medline]
    50. Lin K, Lu H. Why people use social networking sites: An empirical study integrating network externalities and motivation theory. Computers in Human Behavior 2011 May;27(3):1152-1161. [CrossRef]
    51. Lupo D, Erlich Z. Computer literacy and applications via distance e-learning. Computers & Education 2001 May;36(4):333-345. [CrossRef]
    52. McDonald D. Computer Literacy Skills for Computer Information Systems Majors: A Case Study. Journal of Information Systems Education 2004;15(1):19-33.
    53. Debatin B, Lovejoy J, Horn A, Hughes B. Facebook and online privacy: attitudes, behaviors, and unintended consequences. Journal of Computer-Mediated Communication 2009;15(1):83-108. [CrossRef]
    54. Freiberg MS, Cabral HJ, Heeren TC, Vasan RS, Curtis ER, Third National Health and Nutrition Examination Survey. Alcohol consumption and the prevalence of the Metabolic Syndrome in the US: a cross-sectional analysis of data from the Third National Health and Nutrition Examination Survey. Diabetes Care 2004 Dec;27(12):2954-2959. [Medline]
    55. Koch-Weser S, Bradshaw YS, Gualtieri L, Gallagher SS. The Internet as a health information source: findings from the 2007 Health Information National Trends Survey and implications for health communication. J Health Commun 2010;15 Suppl 3:279-293. [CrossRef] [Medline]
    56. Chou WS, Liu B, Post S, Hesse B. Health-related Internet use among cancer survivors: data from the Health Information National Trends Survey, 2003-2008. J Cancer Surviv 2011 Sep;5(3):263-270. [CrossRef] [Medline]
    57. Powell J, Inglis N, Ronnie J, Large S. The characteristics and motivations of online health information seekers: cross-sectional survey and qualitative interview study. J Med Internet Res 2011 Feb;13(1):e20 [FREE Full text] [CrossRef] [Medline]
    58. Livingston G, Minushkin S, Cohn D. Hispanics and Health Care in the United States: Access, Information and Knowledge. Washington, DC: Pew Internet & American Life Project; 2008 Aug 13.   URL: http:/​/www.​pewhispanic.org/​2008/​08/​13/​hispanics-and-health-care-in-the-united-states-access-information-and-knowledge/​ [accessed 2015-04-20] [WebCite Cache]
    59. Webb TL, Joseph J, Yardley L, Michie S. Using the internet to promote health behavior change: a systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. J Med Internet Res 2010;12(1):e4 [FREE Full text] [CrossRef] [Medline]
    60. Siliquini R, Ceruti M, Lovato E, Bert F, Bruno S, De Vito E, et al. Surfing the internet for health information: an italian survey on use and population choices. BMC Med Inform Decis Mak 2011;11:21 [FREE Full text] [CrossRef] [Medline]
    61. McCully SN, Don BP, Updegraff JA. Using the Internet to help with diet, weight, and physical activity: results from the Health Information National Trends Survey (HINTS). J Med Internet Res 2013;15(8):e148 [FREE Full text] [CrossRef] [Medline]
    62. CDC MMWR Weekly. 2007 Mar 16. Fruit and vegetable consumption among adults-United States, 2005   URL: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5610a2.htm [accessed 2015-02-23] [WebCite Cache]
    63. Kalichman SC, Benotsch EG, Weinhardt LS, Austin J, Luke W. Internet use among people living with HIV/AIDS: association of health information, health behaviors, and health status. AIDS Educ Prev 2002 Feb;14(1):51-61. [Medline]
    64. Kalichman SC, Cain D, Cherry C, Pope H, Eaton L, Kalichman MO. Internet use among people living with HIV/AIDS: coping and health-related correlates. AIDS Patient Care STDS 2005 Jul;19(7):439-448. [CrossRef] [Medline]
    65. Samal L, Saha S, Chander G, Korthuis PT, Sharma RK, Sharp V, et al. Internet health information seeking behavior and antiretroviral adherence in persons living with HIV/AIDS. AIDS Patient Care STDS 2011 Jul;25(7):445-449 [FREE Full text] [CrossRef] [Medline]
    66. Shahab L, Brown J, Gardner B, Smith SG. Seeking health information and support online: does it differ as a function of engagement in risky health behaviors? Evidence from the health information national trends survey. J Med Internet Res 2014;16(11):e253 [FREE Full text] [CrossRef] [Medline]
    67. Morello MI, Madanat H, Crespo NC, Lemus H, Elder J. Associations among parent acculturation, child BMI, and child fruit and vegetable consumption in a Hispanic sample. J Immigr Minor Health 2012 Dec;14(6):1023-1029. [CrossRef] [Medline]
    68. Duffey KJ, Gordon-Larsen P, Ayala GX, Popkin BM. Birthplace is associated with more adverse dietary profiles for US-born than for foreign-born Latino adults. J Nutr 2008 Dec;138(12):2428-2435 [FREE Full text] [CrossRef] [Medline]
    69. Sallis JF. Age-related decline in physical activity: a synthesis of human and animal studies. Med Sci Sports Exerc 2000 Sep;32(9):1598-1600. [Medline]
    70. Peltzer K, Pengpid S. Correlates of healthy fruit and vegetable diet in students in low, middle and high income countries. Int J Public Health 2015 Jan;60(1):79-90. [CrossRef] [Medline]
    71. Blisard N, Lin B, Cromartie J, Ballenger N. America’s changing appetite: food consumption and spending to 2020. FoodReview 2002;25(1):2-11.
    72. Parada H, Horton LA, Cherrington A, Ibarra L, Ayala GX. Correlates of medication nonadherence among Latinos with type 2 diabetes. Diabetes Educ 2012;38(4):552-561. [CrossRef] [Medline]
    73. Lanouette NM, Folsom DP, Sciolla A, Jeste DV. Psychotropic medication nonadherence among United States Latinos: a comprehensive literature review. Psychiatr Serv 2009 Feb;60(2):157-174 [FREE Full text] [CrossRef] [Medline]
    74. Shim JW. The use of e-health information and health behavior change: the role of perceived health status and types of e-health information use. TOCOMMJ 2008 Oct 30;2(1):156-163. [CrossRef]
    75. Steptoe A, McMunn A. Health behaviour patterns in relation to hypertension: the English Longitudinal Study of Ageing. J Hypertens 2009 Feb;27(2):224-230. [CrossRef] [Medline]
    76. Shaw RJ, Johnson CM. Health information seeking and social media use on the internet among people with diabetes. Online J Public Health Inform 2011;3(1):pii [FREE Full text] [CrossRef] [Medline]
    77. iHealthBeat. 2011 Mar 21. US patients turn to facebook for health information needs   URL: http:/​/www.​ihealthbeat.org/​articles/​2011/​3/​21/​us-patients-turn-to-facebook-for-health-information-needs [WebCite Cache]
    78. Maher CA, Lewis LK, Ferrar K, Marshall S, De BI, Vandelanotte C. Are health behavior change interventions that use online social networks effective? A systematic review. J Med Internet Res 2014;16(2):e40 [FREE Full text] [CrossRef] [Medline]
    79. Fernandez-Luque L, Karlsen R, Bonander J. Review of extracting information from the Social Web for health personalization. J Med Internet Res 2011;13(1):e15 [FREE Full text] [CrossRef] [Medline]
    80. Wolf MS, Gazmararian JA, Baker DW. Health literacy and health risk behaviors among older adults. Am J Prev Med 2007 Jan;32(1):19-24. [CrossRef] [Medline]
    81. Chisolm DJ, Manganello JA, Kelleher KJ, Marshal MP. Health literacy, alcohol expectancies, and alcohol use behaviors in teens. Patient Educ Couns 2014 Nov;97(2):291-296. [CrossRef] [Medline]
    82. Health Literacy: The Solid Facts. Copenhagen: World Health Organization; 2013.   URL: http://www.euro.who.int/__data/assets/pdf_file/0008/190655/e96854.pdf [WebCite Cache]
    83. Jacobs RJ, Caballero J, Ownby RL, Kane MN. Development of a culturally appropriate computer-delivered tailored Internet-based health literacy intervention for Spanish-dominant Hispanics living with HIV. BMC Med Inform Decis Mak 2014;14:103 [FREE Full text] [CrossRef] [Medline]
    84. van Servellen G, Nyamathi A, Carpio F, Pearce D, Garcia-Teague L, Herrera G, et al. Effects of a treatment adherence enhancement program on health literacy, patient-provider relationships, and adherence to HAART among low-income HIV-positive Spanish-speaking Latinos. AIDS Patient Care STDS 2005 Nov;19(11):745-759. [CrossRef] [Medline]
    85. Koskan A, Friedman DB, Hilfinger Messias DK. Health literacy among Hispanics: a systematic research review (1992–2008). Hisp Hlth Care Int 2010 Jun 01;8(2):65-76. [CrossRef]
    86. Ghaddar SF, Valerio MA, Garcia CM, Hansen L. Adolescent health literacy: the importance of credible sources for online health information. J Sch Health 2012 Jan;82(1):28-36. [CrossRef] [Medline]
    87. Lee K, Hoti K, Hughes JD, Emmerton LM. Interventions to assist health consumers to find reliable online health information: a comprehensive review. PLoS One 2014;9(4):e94186 [FREE Full text] [CrossRef] [Medline]
    88. Berland GK, Elliott MN, Morales LS, Algazy JI, Kravitz RL, Broder MS, et al. Health information on the Internet: accessibility, quality, and readability in English and Spanish. JAMA 2001;285(20):2612-2621 [FREE Full text] [Medline]
    89. Ginossar T, Nelson S. Reducing the health and digital divides: a model for using community-based participatory research approach to e-health interventions in low-income Hispanic communities. Journal of Computer-Mediated Communication 2010;15(4):530-551. [CrossRef]
    90. Utami D, Bickmore TW, Barry B, Paasche-Orlow MK. Health literacy and usability of clinical trial search engines. J Health Commun 2014;19 Suppl 2:190-204. [CrossRef] [Medline]
    91. Atkinson NL, Saperstein SL, Pleis J. Using the internet for health-related activities: findings from a national probability sample. J Med Internet Res 2009;11(1):e4 [FREE Full text] [CrossRef] [Medline]
    92. Bansil P, Keenan NL, Zlot AI, Gilliland JC. Health-related information on the Web: results from the HealthStyles Survey, 2002-2003. Prev Chronic Dis 2006 Apr;3(2):A36 [FREE Full text] [Medline]
    93. Bernstam EV, Shelton DM, Walji M, Meric-Bernstam F. Instruments to assess the quality of health information on the World Wide Web: what can our patients actually use? Int J Med Inform 2005 Jan;74(1):13-19. [CrossRef] [Medline]
    94. Graber MA, D'Alessandro DM, Johnson-West J. Reading level of privacy policies on Internet health Web sites. J Fam Pract 2002 Jul;51(7):642-645. [Medline]
    95. Vann L. MediaPost. 2011 Nov 10. Online Hispanics have a hard time finding health information in Spanish   URL: http:/​/www.​mediapost.com/​publications/​article/​162126/​online-hispanicshave-a-hard-time-finding-health-i.​html [accessed 2015-02-23] [WebCite Cache]
    96. Laz TH, Berenson AB. Racial and ethnic disparities in internet use for seeking health information among young women. J Health Commun 2013;18(2):250-260 [FREE Full text] [CrossRef] [Medline]
    97. Tu HT, Cohen GR. Striking jump in consumers seeking health care information. Track Rep 2008 Aug(20):1-8. [Medline]
    98. Minor M. Commentary – comparing the Hispanic and non‐Hispanic markets: how different are they? Journal of Services Marketing 1992 Feb;6(2):29-32. [CrossRef]
    99. Peña-Purcell N. Hispanics' use of Internet health information: an exploratory study. J Med Libr Assoc 2008 Apr;96(2):101-107 [FREE Full text] [CrossRef] [Medline]
    100. Donaldson S, Grant-Vallone E. Understanding self-report bias in organizational behavior research. Journal of Business and Psychology 2002;17(2):245-260. [CrossRef]
    101. Davis RE, Couper MP, Janz NK, Caldwell CH, Resnicow K. Interviewer effects in public health surveys. Health Educ Res 2010 Feb;25(1):14-26 [FREE Full text] [CrossRef] [Medline]
    102. Lopez MH, Gonzalez-Barrera A, Patten E. Closing the Digital Divide: Latinos and Technology Adoption. Washington, DC: Pew Hispanic Center; 2013 Mar 07.   URL: http://www.pewhispanic.org/files/2013/03/Latinos_Social_Media_and_Mobile_Tech_03-2013_final.pdf [WebCite Cache]

    ‎
    MMAS-8: Morisky 8-Item Medication Adherence Scale
    NYCHANES: New York City Health and Nutrition Examination Survey
    SNS: social networking site
    WICER: Washington Heights Inwood Informatics Infrastructure for Comparative Effectiveness Research

    Edited by G Eysenbach; submitted 22.02.15; peer-reviewed by S Paige; comments to author 12.03.15; revised version received 20.04.15; accepted 07.10.15; published 26.11.15

    Copyright

    ©Young Ji Lee, Bernadette Boden-Albala, Haomiao Jia, Adam Wilcox, Suzanne Bakken. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 26.11.2015.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.