Internet Use and Preventive Health Behaviors Among Couples in Later Life: Evidence from the Health and Retirement Study

Internet Use and Preventive Health Behaviors Among Couples in Later Life: Evidence from the... Abstract Background and Objectives The aim of this study was to examine the link between internet use and preventive health behaviors. We focused on couples to examine whether there were cross-partner associations between internet use and preventive health behaviors. Research Design and Methods The data for this study came from the 2010 and 2012 waves of the Health and Retirement Study and the sample consisted of 5,143 pairs of coupled-individuals. Preventive health behaviors included cancer screenings (mammogram and prostate tests), cholesterol tests, and flu shots. Logistic multilevel actor–partner interdependence models were employed to test the study hypotheses. Results Internet use was associated with a higher likelihood of receiving prostate exams and cholesterol tests for husbands, net of demographic and health characteristics, and insurance status. We found that wives’ internet use was associated with a higher likelihood of receiving flu shots and prostate exams for husbands, but husbands’ internet use was not associated with wives’ preventive health behaviors. Discussion and Implications Research linking internet use and preventive health behaviors is important because such behaviors are associated not only with health of the older population but also with substantial reductions in health care expenditures. Our findings suggested that internet use of older adults is associated with their own preventive health behaviors, as well as their spouses’ preventive health behaviors. Interventions and programs to facilitate older adults’ preventive health behaviors should consider couple-based approaches. Preventive health behaviors, Internet use, Actor–partner interdependence models, Health and Retirement Study Older adults’ use of preventive health care services reduces health risks, affects well-being, and saves lives (Krist et al., 2012). Preventing disease by using health care services not only helps individuals stay healthy, but also has broader positive impacts for the national economy (Centers for Disease Control [CDC], 2017). Researchers have suggested that increasing the use of clinical preventive services could significantly contribute to the health and longevity of older adults in the United States (Farley, Dalal, Mostashari, & Frieden, 2010). For example, cancer screenings such as mammograms and prostate tests when preformed in recommended time periods have been shown to reduce mortality (Centers for Disease Control and Prevention [CDC], Administration on Aging [AOA], Agency for Healthcare Research and Quality [AHRQ], & Centers for Medicare and Medicaid Services [CMS], 2011). Despite the effectiveness of these preventive health services, less than 50% of older Americans aged 65 or older used these services; and even individuals whose insurance provided partial or full coverage for these services received no services or fewer services than for which they were eligible (CDC, 2017; Krist et al., 2012). Specifically, eliminating costs did not seem to significantly increase the use of preventive health care services among older adults (Ozminkowski et al., 2006). To address the chronic disease burden of both individuals and the larger society, various interventions have been implemented and tested (Bauer, Briss, Goodman, & Bowman, 2014). However, broad public support to promote preventive health behaviors and efforts from multiple initiatives have not been highly effective to increase the delivery of preventive health care services (Krist et al., 2012). The literature on the use of preventive health care services of older adults suggested that logistical challenges, lack of knowledge, limited information and motivation, and insufficient collaboration of local and national agencies may work as barriers (Krist et al., 2012; National Prevention Council [NPC], 2011). A common strategy underlying recent efforts to facilitate the use of preventive health care services is providing health-related information using the internet (CDC, AOA, AHRQ, & CMS, 2011; NPC, 2011). Although the link between internet use and access to health information has been examined, prior research has mostly focused on the age related digital health divide (Hall, Bernhardt, Dodd, & Vollrath, 2015; Koch-Weser, Bradshaw, Gualtieri, & Gallagher, 2010; Moorhead et al., 2013), possible harm caused by inaccurate health information on the web (Chesser, Burke, Reyes, & Rohrberg, 2016; Cole, Watkins, & Kleine, 2016), and intervention programs using the internet (Bennett & Glasgow, 2009; Payne, Lister, West, & Bernhardt, 2015; Widmer et al., 2015). Findings from previous studies suggest that the internet can be a source of online health consultations and elimination of false beliefs about certain health conditions (Lo, Esser, & Gordon, 2010; Lu, Shaw, & Gustafson, 2011). However, less is known about of the link between older adults’ internet use and preventive health care behaviors. Further, even less is known about the couple dynamics underlying the link between internet use and preventive health care behaviors (Webb, Joseph, Yardley, & Michie, 2010; Xavier et al., 2013). Most studies to date focused on individual characteristics associated with preventive health care behaviors; this is surprising, given that spouses represent the most important source of care and medical information across the life course (Meyler, Stimpson, & Peek, 2007; Smith, 2011). Findings from earlier studies are also limited by small, non-representative samples and cross-sectional designs (Amante, Hogan, Pagoto, English, & Lapane, 2015; Ayers & Kronenfeld, 2007; Xavier et al., 2013). In this study, we addressed some of these shortcomings by using nationally representative data from the Health and Retirement Study and examining the association between internet use and subsequent preventive health behaviors more than a 2-year study period. Further, we examined whether there are cross-partner associations between internet use and preventive health behaviors. Older Adults’ Internet Use and Preventive Health Behaviors The internet has become an important source of health-related information over the past two decades (Brodie et al., 2000). The expansion of technology has resulted in a rapid increase in the use of the internet to seek health information (Hayward, Hummer, & Sasson, 2015). In addition, a main reason older adults use the internet is to seek health-related information (Heart & Kalderon, 2013). In particular, people who have limited health care services were more likely to use the internet to obtain health information (Amante et al., 2015). Research literature also indicates that internet use influences positive health behaviors such as better diet, daily exercise, and smoking abstinence (Redmond, Baer, Clark, Lipsitz, & Hicks, 2010; Shahab, Brown, Gardner, & Smith, 2014; Xavier et al., 2013). Other studies have shown that the internet provides an accessible resource for obtaining health information (Kivits, 2009), and older adults have also been shown to refer to the internet for health information once they identify a trustworthy source (e.g., WebMD; Walker et al., 2017). Older adults’ acquisition of health information on the internet may influence their health behaviors. As such, health care professionals have been working to develop various web-based interventions and programs to educate older adults with accurate and appropriate health information (Calvillo, Román, & Roa, 2013). Taken together, these findings suggest that internet use may also be associated with other forms of health behaviors, but the link between internet use and preventive health care behaviors remains an understudied area of research. One study showed that older internet users in the United Kingdom were more likely to receive colorectal cancer screenings, but internet use was not associated with receiving mammograms (Xavier et al., 2013). In sum, this literature provides a basis for our hypothesis that internet use would be positively associated with preventive health behaviors among individuals (i.e., actor effects). Older Couples’ Internet Use and Preventive Health Behaviors Life course scholarship has long emphasized the importance of family relationships for well-being in later life. In particular, a substantial body of work has highlighted the principle of linked lives to emphasize the interconnectedness of family relationships and their impact for health and well-being (Elder, Johnson, & Crosnoe, 2003; Gilligan et al., 2017). In this context, the marital dyad is considered to be one of the most important family relationships affecting individuals in later life (Thomas, Liu, & Umberson, 2017). Marital relationships are not only understood to confer significant health benefits, but the health of individuals in marital dyads is also known to be concordant (Meyler et al., 2007; Pai, Godboldo-Brooks, & Edington, 2010). Such health concordance is in part attributable to shared information and health behaviors among couples, as well as the social control function that spouses exert on each other (Thomas et al., 2017). For example, health beliefs and behaviors of a spouse have been shown to influence the other partners’ health behaviors (Manne, Kashy, Weinberg, Boscarino, & Bowen, 2012). Others have shown that marital satisfaction predicts use of cancer screenings among older couples (Kotwal, Lauderdale, Waite, & Dale, 2016). Marriage predicted older men’s colonoscopy use, but not their partners’ use, suggesting some cross-spousal influences in preventive health care behavior (Kotwal et al., 2016). Due to the interdependence of married couples in later life, it is possible that one spouse’s internet use could affect the preventive health behaviors of the other spouse. Therefore, we used a sample of older coupled-individuals to examine cross-partner associations between internet use and preventive health behaviors. Other Factors Affecting Older Adult’s Internet Use and Preventive Health Behaviors Previous literature has demonstrated associations between factors such as race/ethnicity, education, income, and employment status and preventive health care service utilization. For example, older, low-income and ethnic-minority Americans were disproportionately less likely to use these services (CDC, AOA, AHRQ, & CMS, 2011). However, age, education, and race/ethnicity were not significant factors affecting colonoscopy use, whereas higher income was associated with this use (Kotwal et al., 2016). History of medical conditions was also related with preventive health behaviors (Kotwal et al., 2016; Manne et al., 2012). Taken together, this work indicates that certain populations are less likely to use preventive health care services; however, the mechanisms explaining these disparities are somewhat equivocal and not well understood (CDC, AOA, AHRQ, & CMS, 2011; Ozminkowski et al., 2006). Therefore, we will take these demographic factors into consideration when examining the association between older adult’s internet use and preventive health behaviors. In sum, the purpose of this study is to examine the association between internet use and preventive health care behaviors among older couples. Using nationally representative household data for coupled-individuals in the Health and Retirement Study, we investigate whether internet use of older adults and their spouses are prospectively associated with the use of preventive health care services, including influenza vaccinations and cancer screenings, over a 2-year period. Based on the empirical findings discussed previously, we hypothesized that internet use would be positively associated with preventive health care behavior among individuals. We also hypothesized that there would be cross-partner relationships between internet use and preventive health care behavior. Data and Methods Data Source and Study Sample This study was based on the nationally representative Health and Retirement Study (HRS), a biennial longitudinal panel study of individuals over age 51 and their spouses (of any age) in the United States. The HRS, which was initiated in 1992, has collected information from more than 23,000 households (Sonnega et al., 2014). Most of the data for this study came from a file constructed by the Rand Corporation (Version P; Bugliari et al., 2016); measures for internet use came from the publicly available files provided by the Institute of Social Research at the University of Michigan. The key objective of this study was to examine whether internet use was prospectively associated with preventive health care service use. Although the HRS is fielded every 2 years, data on preventive health care are collected every other wave (i.e., every 4 years), with the most recent data available from the 2012 wave. Therefore, we identified the 2010 wave as the baseline point, from which information for the key independent measure (i.e., internet use) and other covariates were assessed (for an exception, see subsequent section on measures for previous preventive health care service utilization); information for preventive health care service utilization was taken from the 2012 wave, which is the follow-up directly following the baseline. The study sample consisted of coupled-individuals who were in a heterosexual marriage (92%) or an otherwise similar partnership (8%). (Same-sex couples were excluded because the analytic technique used in this study is applicable to “distinguishable” dyads only; Kenny & Ledermann, 2010.) The coupled-individuals in the sample were also (a) non-proxy respondents at baseline; and (b) re-interviewed at follow-up. Among couples where both spouses satisfied these criteria (dyad N = 5,188), very few couples had missing information on study variables. That is, 11 couples who had missing information for the internet use measures and 34 couples who had missing information for other study measures were excluded from the sample (0.8%). List-wise deletion was used to handle missing data because there were fewer than 1% missing on any variable in the analysis (cf. Allison, 2010). The final sample included 5,143 coupled-individuals (individual N = 10,286). To ensure that health care service was “preventive” in nature, we applied additional sampling criteria for analyzing specific health care outcomes (Kim & Kawachi, 2017). For example, couples where both spouses did not have a history of stroke or heart disease were selected for the analysis of cholesterol tests (dyad n = 3,132); similarly, couples where both spouses did not have a history of cancer were selected for the analysis of cancer screening (e.g., mammograms and prostate cancer screening; dyad n = 3,958). Of the respective samples, a small fraction of couples had missing information on the outcome variable for one of the spouses (influenza vaccination [dyad n = 12; 0.2%], cholesterol tests [dyad n = 40; 1.3%], and cancer screenings [dyad n = 96; 2.4%]). Because the multilevel analytic approach taken in this study facilitated analysis of couples where a spouse had missing information on the outcome variable (Loeys & Molenberghs, 2013), these couples were retained in the analyses; estimates from sensitivity analyses excluding these couples were consistent with the main findings (results available upon request). Measures Preventive Health Care Behaviors At follow-up, respondents were asked to report on preventive health care services that they received since they were last interviewed (at baseline) with the question: “In the last two years, have you had any of the following medical tests or procedures?” (a) a flu shot, (b) a blood test for cholesterol, (c) a mammogram or x-ray of the breast to search for cancer (for women), and (d) a prostate-specific antigen (PSA) blood test or other examination to screen for prostate cancer (for men). Responses to items regarding flu shots and cholesterol tests were coded dichotomously (1 = yes; 0 = no). A measure for sex-specific cancer screenings was created based on whether the wife and husband had screened for breast and prostate cancer, respectively, which was also coded dichotomously. Information on other preventive health care services were also available from the 2012 wave of the HRS, which included a colonoscopy, sigmoidoscopy, or other screening for colon cancer and a pap smear for cervical cancer. We decided not to include colon cancer screening in our main analysis because 2012 was the first year in which the measure was obtained and no information on previous use was available; an exploratory analysis pertaining to the link between internet use and colon cancer screening is provided in a supplementary material (Supplementary Table 1). Information on Pap smear was also used in a supplementary analysis where female-specific cancer screening was defined as breast cancer or cervical cancer screenings (Supplementary Table 2). Internet Use Respondents’ internet use was assessed at baseline with the following question “Do you regularly use the World Wide Web, or the Internet, for sending and receiving e-mail or for any other purpose, such as making purchases, searching for information, or making travel reservations?” The response to this single-item measure was coded dichotomously (1 = yes; 0 = no). Covariates The HRS contains information on an extensive set of covariates documented or postulated to affect preventive health care behavior, which were considered in the analyses. Demographic characteristics included age (in years), race/ethnic status (non-Hispanic White [reference], non-Hispanic Black, non-Hispanic “other” race, and Hispanic), household income (transformed by the natural log in the analyses), and education (no educational degree [reference], high school diploma or GED, some college, college degree or higher). Health care and health characteristics included insurance coverage through Medicare or Medicaid, either spouse’s employer (current or past), or any other supplemental source (1 = insured; 0 = not insured), whether respondents had seen or talked to medical doctors or nurse practitioners about health over the past 2 years (1 = visited doctor; 0 = no), and number of chronic conditions ever diagnosed by a clinician, including (a) high blood pressure or hypertension, (b) diabetes, (c) cancer (not included for the analyses of cancer screenings), (d) lung disease, (e) heart conditions, (f) stroke, (g) psychiatric problems, and (h) arthritis or rheumatism (range = 0–8). Because preventive health care utilization (i.e., flu shot, cholesterol test, or sex-specific cancer screening) was not assessed at baseline (2010), we created a measure for whether HRS records indicated that respondents had ever received the respective screening or test (1 = used preventive health care; 0 = no). Marriage-related covariates included marital status (1 = married; 0 = non-marital cohabitation), first marriage (1 = first marriage; 0 = others), parental status (1 = has any child(ren); 0 = no child), and length of marriage in years. All covariates except previous preventive health care behavior were measured at baseline. Analytic Strategy The relationship between internet use and subsequent preventive health care utilization during the 2-year observation period was estimated using the multilevel actor–partner interdependence model (APIM) framework widely used to analyze dyadic data (Kenny & Ledermann, 2010). Given that preventive health care service utilization measures were binary, we used generalized estimating equations with an extension for logistic regression models (PROC GEMNOD, SAS Version 9.4; Loeys, Cook, de Smet, Wietzker, & Buysse, 2014). To examine whether internet use was related to preventive health care utilization for both wives and husbands while allowing for the non-independence of the preventive health care utilization within couples, we employed a two-intercept model to estimate the associations. Importantly, the association between one’s own internet use (i.e., actor effects; Hypothesis 1), as well as that of the spouse’s (i.e., partner effects; Hypothesis 2), and preventive health care utilization were simultaneously estimated (Loeys et al., 2014). For each preventive health care behavior, we present two sets of APIMs to test actor and partner effects of internet use. Model 1 included actor and partner effects of internet use without covariate adjustment (i.e., unadjusted model); Model 2 introduced the full set of covariates (i.e., adjusted model). Results Study sample characteristics are presented in Table 1. Approximately 65% of wives and husbands had received an influenza vaccination during the 2-year observation period (gender differences not statistically significant). More than 80% of the sample without a history of a heart disease or stroke at baseline had received a cholesterol test, with wives showing a slightly higher rate compared with husbands. With regard to sex-specific cancer screenings, approximately 72% of wives and 63% of husbands without a history of cancer at baseline had received screenings for breast and prostate cancer, respectively. A majority of the study sample reported using the internet, with wives (62%) showing a higher usage rate compared with husbands (56%). Wives were generally healthier compared with husbands, as indicated by the number of chronic health conditions. Approximately 92% of the couples in the sample were in a formal marriage (as opposed to being in a non-marital partnership). Couples were married for an average of about 30 years. Table 1. Descriptive Characteristics of the Health and Retirement Study Sample   Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —      Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —    Notes: Dyad N = 5,143. M = mean. SD = standard deviation. Group differences were tested using t tests for continuous variables and the chi-square statistics for categorical variables. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). cWhether previous HRS records (1996–2008) indicate preventive health care use. dCouple-level covariates. eInsurance coverage through Medicare/Medicaid, either spouse’s employer (current or past), or any other supplemental insurance. fCount of eight chronic health conditions. gMarital status (1 = married; 0 = non-marital cohabitation). *p < .05. **p < .01. ***p < .001. View Large Table 1. Descriptive Characteristics of the Health and Retirement Study Sample   Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —      Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —    Notes: Dyad N = 5,143. M = mean. SD = standard deviation. Group differences were tested using t tests for continuous variables and the chi-square statistics for categorical variables. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). cWhether previous HRS records (1996–2008) indicate preventive health care use. dCouple-level covariates. eInsurance coverage through Medicare/Medicaid, either spouse’s employer (current or past), or any other supplemental insurance. fCount of eight chronic health conditions. gMarital status (1 = married; 0 = non-marital cohabitation). *p < .05. **p < .01. ***p < .001. View Large Influenza Vaccination Table 2 contains results from multilevel models predicting preventive health care use in relation to one’s own and partner’s internet use. The rows containing estimates of actor and partner effects of internet use address Hypotheses 1 and 2, respectively. In the models predicting influenza vaccination, one’s own internet use was not associated with vaccine uptake for either spouse in the unadjusted and adjusted models. Partner’s internet use was also not associated with one’s own vaccine uptake for either spouse in the unadjusted model. However, the partner effect became significant for husbands, but not wives, when covariates were introduced to the model; that is, husbands were more likely to obtain influenza vaccination when their wives used the internet compared with when wives did not use the internet (adjusted OR = 1.22, p < .05). Table 2. Odds Ratios and 95% Confidence Intervals for the Relationship Between Internet Use (2010) and Preventative Health Behaviors (2012)     Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]      Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]  Notes: Models fully adjusted for age, race/ethnic status, education, household income, insurance status, doctor visits, number of chronic health conditions, previous preventive health care utilization, marital status, whether the current marriage is the first marriage, parental status, and length of marriage in years. Dyad N = 5,143; 95% confidence intervals were presented in square brackets. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). *p < .05. **p < .01. ***p < .001. View Large Table 2. Odds Ratios and 95% Confidence Intervals for the Relationship Between Internet Use (2010) and Preventative Health Behaviors (2012)     Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]      Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]  Notes: Models fully adjusted for age, race/ethnic status, education, household income, insurance status, doctor visits, number of chronic health conditions, previous preventive health care utilization, marital status, whether the current marriage is the first marriage, parental status, and length of marriage in years. Dyad N = 5,143; 95% confidence intervals were presented in square brackets. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). *p < .05. **p < .01. ***p < .001. View Large Cholesterol Test In the unadjusted model, actor effects of internet use were associated with receiving cholesterol tests for both spouses, such that wives (unadjusted OR = 1.31, p < .05) and husbands (unadjusted OR = 1.80, p < .001) who reported using the internet at baseline were more likely to screen for cholesterol compared with those who did not use the internet. In the model fully adjusted for covariates, the actor effect remained significant for husbands (adjusted OR = 1.51, p < .01), but not wives (adjusted OR = 1.16, p = .28). Further, wives’ internet use was significantly associated with husbands’ cholesterol test (unadjusted OR = 1.33, p < .01) in the unadjusted model, but this partner effect was no longer statistically significant in the fully adjusted model (adjusted OR = 1.12, p = .43). Husbands’ internet use was not associated with wives’ cholesterol tests in the unadjusted and adjusted models. Cancer Screening Wives’ and husbands’ internet use was associated with increased likelihood of receiving breast cancer screening for wives (unadjusted OR = 1.48, p < .001) and prostate cancer screening for husbands (unadjusted OR = 1.69, p < .001) in the unadjusted model. The actor effects of internet use remained significant in terms of husbands’ prostate cancer screening in the fully adjusted model (adjusted OR = 1.38, p < .001), but actor effects became marginally significant in the fully adjusted model with regard to wives’ breast cancer screening (adjusted OR = 1.23, p = .052). Also, wives’ internet use was significantly associated with husbands’ prostate cancer screening in the unadjusted (unadjusted OR = 1.34, p < .001), but the effects became marginally significant in the fully adjusted (adjusted OR = 1.20, p = .052) model. Husbands’ internet use was significantly associated with wives’ breast cancer screening in the unadjusted (unadjusted OR = 1.30, p < .01), but the effects were no longer significant in the fully adjusted model. Additional analyses were performed to examine whether the gendered pattern of our findings was statistically significant, using an alternative parameterization of the two-intercept model, known as the “interaction approach” (Loeys et al., 2014). However, results from these additional models indicated that the gender differences in the association between internet use and preventive health care behavior did not reach statistical significance (results available upon request). Discussion Given the disparity in the use of preventive health care services in later life, our central aim in this article was to understand how internet use of older adults and their partners plays a role in preventive health behaviors. Drawing from the life course perspective, which emphasizes the concept of “linked lives,” we examined whether older couples’ internet use was associated with preventive health behaviors over a 2-year period. Specifically, we explored the association between older adults’ internet use and their preventive health behaviors as well as their partners’ internet use in these associations. Overall, the findings from this study indicate that both older adults’ internet use and their partners’ internet use promoted preventive health behaviors. This was true across the range of preventive health behaviors considered, including flu shots, cholesterol tests, mammograms, and prostate exams. We expected that one’s own use of the internet would be positively associated with preventive health behaviors (Hypothesis 1). In regard to cholesterol tests and cancer screenings, our findings supported this hypothesis. In the unadjusted model, internet use was associated with both husbands’ and wives’ receipt of cholesterol and cancer screenings. In the model fully adjusted for covariates (i.e., sociodemographic characteristics, health and health care characteristics, and couple characteristics), effects of husbands’ own internet use remained significant, but the associations between wives’ internet use and preventive health behaviors were no longer statistically significant. Based on the interdependence of couples in later life, we expected cross-partner effects of internet use for both men and women (Hypothesis 2). The same trend we discovered among individuals’ effects of internet use was found between partners. Wives’ internet use was associated with husbands’ receiving flu shots. However, the effects of husbands’ internet use on wives’ preventive health behaviors attenuated when we considered other factors such as education, experience of health care, and couple characteristics. This finding is consistent with previous research, which has demonstrated that women are more knowledgeable on health and health care throughout life course as result of their reproductive health care needs and care-provision responsibilities (Beier & Ackerman, 2003; Chylińska et al., 2017). Based on these needs and roles, women have more active attitudes toward receiving medical treatments (Chylińska et al., 2017). In addition, women are more likely than men to have early and frequent exposures to health care services (Kent et al., 2017). As a result, some health-related information from the internet may be less relevant to women’s own preventive health behaviors. In line with previous research, which has demonstrated that married people influence their partners’ health in later-life (Meyler et al., 2007; Pai et al., 2010), our findings indicate that older adults’ internet use is associated with their own preventive health behaviors, as well as their spouses’ preventive health behaviors. However, given that the gender differences in the association between wives and husbands in the sample did not reach statistical significance, more research is needed to corroborate the gendered pattern of findings reported in this study. Future Directions The internet use measure included in this study assessed the general use of the internet and is not specific to information acquisition in relation to preventive health behaviors. Thus, the potential pathways linking internet use and preventive health behaviors were not directly evaluated because this information was not available in the Health and Retirement Study. Future research should consider these associations between health-related internet use and preventative health behaviors. Second, a strength of this study was its ability to examine the associations between internet use and preventive health behaviors among older couples. However, recent research indicates the growing prevalence of alternative union forms (Benson & Coleman, 2016; Fredriksen-Goldsen, Hoy-Ellis, Goldsen, Emlet, & Hooyman, 2014). Future research should consider whether older adults in these alternative union forms also receive health benefits from internet use. Further, because the study focused on couples, our findings are not generalizable to older individuals who are not in relationships. Given that marriage has health benefits for older adults (Meyler et al., 2007; Pai et al., 2010), single older adults are particularly vulnerable to health problems. Implications for Practice Taken together, the findings in this study support the documented positive impacts of older adults’ internet use on their health behaviors by specifically highlighting the role internet use has in older adult use of preventative health care services (Bauer et al., 2014; Cheston, Flickinger, & Chisolm, 2013; Ventola, 2014). Research suggests that internet use is not only a source of health-related information, but it can also be an effective intervention strategy to promote health among older adults (Cheston et al., 2013). Specifically, professionals working on the development of interventions and programs to promote preventive health behaviors have highlighted the interactive features of the internet, and web-based programs are expanding rapidly (CDC, 2011; Cheston et al., 2013; Hall et al., 2015). Thus, there is growing interest in developing programs and interventions to promote older adults’ health behaviors. The findings from this study indicated that internet use may be a particularly promising approach to increase older adults’ use of preventative health care services and should be considered in future intervention efforts. Further, given our findings indicating that older adults benefit not just from their own internet use but also their spouses, it may be particularly advantageous to include both partners in programming efforts. Current efforts to increase older adults’ use of preventive health care services have focused almost exclusively on individuals (Arden-Close & McGrath, 2017). Our findings suggest that programs that incorporate dyadic designs may be particularly promising to facilitate older adults’ internet use and engagement in preventative health behaviors (Arden-Close & McGrath, 2017). Supplementary Data Supplementary data are available at The Gerontologist online. References Allison, P. ( 2010). Missing data. In J. Wright & P. Marsden (Eds.), Handbook of survey research  (pp. 631– 658). Bingley, UK: Emerald Group Limited. Amante, D. J., Hogan, T. P., Pagoto, S. L., English, T. M., & Lapane, K. L. ( 2015). Access to care and use of the Internet to search for health information: Results from the US National Health Interview Survey. Journal of Medical Internet Research , 17, e106. doi: 10.2196/jmir.4126 Google Scholar CrossRef Search ADS   Arden-Close, E., & McGrath, N. ( 2017). Health behaviour change interventions for couples: A systematic review. British Journal of Health Psychology , 22, 215– 237. doi: 10.1111/bjhp.12227 Google Scholar CrossRef Search ADS   Ayers, S. L., & Kronenfeld, J. J. ( 2007). Chronic illness and health-seeking information on the Internet. Health (London, England: 1997) , 11, 327– 347. doi: 10.1177/1363459307077547 Google Scholar CrossRef Search ADS   Bauer, U. E., Briss, P. A., Goodman, R. A., & Bowman, B. A. ( 2014). Prevention of chronic disease in the 21st century: Elimination of the leading preventable causes of premature death and disability in the USA. Lancet (London, England) , 384, 45– 52. doi: 10.1016/S0140-6736(14)60648-6 Google Scholar CrossRef Search ADS   Beier, M. E., & Ackerman, P. L. ( 2003). Determinants of health knowledge: An investigation of age, gender, abilities, personality, and interests. Journal of Personality and Social Psychology , 84, 439– 448. doi:10.1037/0022-3514.84.2.439 Google Scholar CrossRef Search ADS   Bennett, G. G., & Glasgow, R. E. ( 2009). The delivery of public health interventions via the Internet: Actualizing their potential. Annual Review of Public Health , 30, 273– 292. doi: 10.1146/annurev.publhealth.031308.100235 Google Scholar CrossRef Search ADS   Benson, J. J., & Coleman, M. ( 2016). Older adults developing a preference for living apart together. Journal of Marriage and Family , 78, 797– 812. doi: 10.1111/jomf.12292 Google Scholar CrossRef Search ADS   Brodie, M., Flournoy, R. E., Altman, D. E., Blendon, R. J., Benson, J. M., & Rosenbaum, M. D. ( 2000). Health information, the Internet, and the digital divide. Health affairs (Project Hope) , 19, 255– 265. doi:10.1377/hlthaff.19.6.255 Google Scholar CrossRef Search ADS   Bugliari, D., Campbell, N., Chan, C., Hayden, O., Hurd, M., Main, R., … St. Clair, P. ( 2016). Rand HRS data documentation, version p . Santa Monica, CA: RAND Center for the Study of Aging, with funding from the National Institute on Aging and the Social Security Administration. Calvillo, J., Román, I., & Roa, L. M. ( 2013). Empowering citizens with access control mechanisms to their personal health resources. International Journal of Medical Informatics , 82, 58– 72. doi: 10.1016/j.ijmedinf.2012.02.006 Google Scholar CrossRef Search ADS   Centers for Disease Control.( 2017). Preventive health care . Retrieved from https://www.cdc.gov/healthcommunication/toolstemplates/entertainmented/tips/PreventiveHealth.html Centers for Disease Control and Prevention, Administration on Aging, Agency for Healthcare Research and Quality, & Centers for Medicare and Medicaid Services. ( 2011). Enhancing use of clinical preventive services among older adults . Washington, DC: AARP. PubMed PubMed  Chesser, A., Burke, A., Reyes, J., & Rohrberg, T. ( 2016). Navigating the digital divide: A systematic review of eHealth literacy in underserved populations in the United States. Informatics for Health & Social Care , 41, 1– 19. doi: 10.3109/17538157.2014.948171 Google Scholar CrossRef Search ADS   Cheston, C. C., Flickinger, T. E., & Chisolm, M. S. ( 2013). Social media use in medical education: A systematic review. Academic Medicine: Journal of the Association of American Medical Colleges , 88, 893– 901. doi: 10.1097/ACM.0b013e31828ffc23 Google Scholar CrossRef Search ADS   Chylińska, J., Łazarewicz, M., Rzadkiewicz, M., Adamus, M., Jaworski, M., Haugan, G., … Włodarczyk, D. ( 2017). The role of gender in the active attitude toward treatment and health among older patients in primary health care–self-assessed health status and sociodemographic factors as moderators. BMC Geriatrics , 17, 284. doi: 10.1186/s12877-017-0677-z Google Scholar CrossRef Search ADS   Cole, J., Watkins, C., & Kleine, D. ( 2016). Health advice from internet discussion forums: How bad is dangerous? Journal of Medical Internet Research , 18, e4. doi: 10.2196/jmir.5051 Google Scholar CrossRef Search ADS   Elder, G. H. Jr., Johnson, M. K., & Crosnoe, R. ( 2003). The emergence and development of life course theory. In J. T. Mortimer & M. J. Shanahan (Eds.), Handbook of the life course  (pp. 3– 19). New York, NY: Springer. Google Scholar CrossRef Search ADS   Farley, T. A., Dalal, M. A., Mostashari, F., & Frieden, T. R. ( 2010). Deaths preventable in the U.S. by improvements in use of clinical preventive services. American Journal of Preventive Medicine , 38, 600– 609. doi: 10.1016/j.amepre.2010.02.016 Google Scholar CrossRef Search ADS   Fredriksen-Goldsen, K. I., Hoy-Ellis, C. P., Goldsen, J., Emlet, C. A., & Hooyman, N. R. ( 2014). Creating a vision for the future: Key competencies and strategies for culturally competent practice with lesbian, gay, bisexual, and transgender (LGBT) older adults in the health and human services. Journal of Gerontological Social Work , 57( 2–4), 80– 107. doi: 10.1080/01634372.2014.890690 Google Scholar CrossRef Search ADS   Gilligan, M., J. J. Suitor, Nam, S., Routh, B., Rurka, M., & Con, G. ( 2017). Family networks and psychological well-being in midlife. Social Sciences , 6, 94. doi: 10.3390/socsci6030094 Google Scholar CrossRef Search ADS   Hall, A. K., Bernhardt, J. M., Dodd, V., & Vollrath, M. W. ( 2015). The digital health divide: Evaluating online health information access and use among older adults. Health Education & Behavior: The Official Publication of the Society for Public Health Education , 42, 202– 209. doi: 10.1177/1090198114547815 Google Scholar CrossRef Search ADS   Hayward, M. D., Hummer, R. A., & Sasson, I. ( 2015). Trends and group differences in the association between educational attainment and U.S. adult mortality: Implications for understanding education’s causal influence. Social Science & Medicine (1982) , 127, 8– 18. doi: 10.1016/j.socscimed.2014.11.024 Google Scholar CrossRef Search ADS   Heart, T., & Kalderon, E. ( 2013). Older adults: Are they ready to adopt health-related ICT? International Journal of Medical Informatics , 82, e209– e231. doi: 10.1016/j.ijmedinf.2011.03.002 Google Scholar CrossRef Search ADS   Kenny, D. A., & Ledermann, T. ( 2010). Detecting, measuring, and testing dyadic patterns in the actor-partner interdependence model. Journal of Family Psychology: JFP , 24, 359– 366. doi: 10.1037/a0019651 Google Scholar CrossRef Search ADS   Kent, S., Fusco, F., Gray, A., Jebb, S. A., Cairns, B. J., & Mihaylova, B. ( 2017). Body mass index and healthcare costs: A systematic literature review of individual participant data studies. Obesity Reviews: An Official Journal of the International Association for the Study of Obesity , 18, 869– 879. doi: 10.1111/obr.12560 Google Scholar CrossRef Search ADS   Kim, E. S., & Kawachi, I. ( 2017). Perceived neighborhood social cohesion and preventive healthcare use. American Journal of Preventive Medicine , 53, e35– e40. doi: 10.1016/j.amepre.2017.01.007 Google Scholar CrossRef Search ADS   Kivits, J. ( 2009). Everyday health and the internet: A mediated health perspective on health information seeking. Sociology of Health & Illness , 31, 673– 687. doi: 10.1111/j.1467-9566.2008.01153.x Google Scholar CrossRef Search ADS   Koch-Weser, S., Bradshaw, Y. S., Gualtieri, L., & Gallagher, S. S. ( 2010). The Internet as a health information source: Findings from the 2007 Health Information National Trends Survey and implications for health communication. Journal of Health Communication , 15( Suppl. 3), 279– 293. doi: 10.1080/10810730.2010.522700 Google Scholar CrossRef Search ADS   Kotwal, A. A., Lauderdale, D. S., Waite, L. J., & Dale, W. ( 2016). Differences between husbands and wives in colonoscopy use: Results from a national sample of married couples. Preventive Medicine , 88, 46– 52. doi: 10.1016/j.ypmed.2016.03.011 Google Scholar CrossRef Search ADS   Krist, A., Shenson, D., Woolf, S., Bradley, C., Liaw, W., & Rothemich, S. ( 2012). A framework for integration of community and clinical care to improve the delivery of clinical preventive services among older adults . National Association of Chronic Disease Directors and the Michigan Public Health Institute. Retrieved from https://www.cdc.gov/aging/pdf/cps-integration-framework.pdf Lo, A. S., Esser, M. J., & Gordon, K. E. ( 2010). YouTube: A gauge of public perception and awareness surrounding epilepsy. Epilepsy & Behavior: E&B , 17, 541– 545. doi: 10.1016/j.yebeh.2010.02.004 Google Scholar CrossRef Search ADS   Loeys, T. O. M., Cook, W., de Smet, O., Wietzker, A., & Buysse, A. N. N. ( 2014). The actor-partner interdependence model for categorical dyadic data: A user-friendly guide to GEE. Personal Relationships , 21, 225– 241. doi: 10.1111/pere.12028 Google Scholar CrossRef Search ADS   Loeys, T., & Molenberghs, G. ( 2013). Modeling actor and partner effects in dyadic data when outcomes are categorical. Psychological Methods , 18, 220– 236. doi: 10.1037/a0030640 Google Scholar CrossRef Search ADS   Lu, H. Y., Shaw, B. R., & Gustafson, D. H. ( 2011). Online health consultation: Examining uses of an interactive cancer communication tool by low-income women with breast cancer. International Journal of Medical Informatics , 80, 518– 528. doi: 10.1016/j.ijmedinf.2011.03.011 Google Scholar CrossRef Search ADS   Manne, S., Kashy, D., Weinberg, D. S., Boscarino, J. A., & Bowen, D. J. ( 2012). Using the interdependence model to understand spousal influence on colorectal cancer screening intentions: A structural equation model. Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine , 43, 320– 329. doi: 10.1007/s12160-012-9344-y Google Scholar CrossRef Search ADS   Meyler, D., Stimpson, J. P., & Peek, M. K. ( 2007). Health concordance within couples: A systematic review. Social Science & Medicine (1982) , 64, 2297– 2310. doi: 10.1016/j.socscimed.2007.02.007 Google Scholar CrossRef Search ADS   Moorhead, S. A., Hazlett, D. E., Harrison, L., Carroll, J. K., Irwin, A., & Hoving, C. ( 2013). A new dimension of health care: Systematic review of the uses, benefits, and limitations of social media for health communication. Journal of Medical Internet Research , 15, e85. doi: 10.2196/jmir.1933 Google Scholar CrossRef Search ADS   National Prevention Council.( 2011). National prevention strategy . Washington, DC: U.S. Department of Health and Human Services, Office of the Surgeon General. Ozminkowski, R. J., Goetzel, R. Z., Shechter, D., Stapleton, D. C., Baser, O., & Lapin, P. ( 2006). Predictors of preventive service use among Medicare beneficiaries. Health Care Financing Review , 27, 5– 23. Pai, C. W., Godboldo-Brooks, A., & Edington, D. W. ( 2010). Spousal concordance for overall health risk status and preventive service compliance. Annals of Epidemiology , 20, 539– 546. doi: 10.1016/j.annepidem.2010.03.020 Google Scholar CrossRef Search ADS   Payne, H. E., Lister, C., West, J. H., & Bernhardt, J. M. ( 2015). Behavioral functionality of mobile apps in health interventions: A systematic review of the literature. JMIR Mhealth and Uhealth , 3, e20. doi: 10.2196/mhealth.3335 Google Scholar CrossRef Search ADS   Redmond, N., Baer, H. J., Clark, C. R., Lipsitz, S., & Hicks, L. S. ( 2010). Sources of health information related to preventive health behaviors in a national study. American Journal of Preventive Medicine , 38, 620– 627.e2. doi: 10.1016/j.amepre.2010.03.001 Google Scholar CrossRef Search ADS   Shahab, L., Brown, J., Gardner, B., & Smith, S. G. ( 2014). 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. Journal of Medical Internet Research , 16, e253. doi: 10.2196/jmir.3368 Google Scholar CrossRef Search ADS   Smith, D. ( 2011). Health care consumer’s use and trust of health information sources. Journal of Communication in Healthcare , 4, 200– 210. doi: 10.1179/1753807611Y.0000000010 Google Scholar CrossRef Search ADS   Sonnega, A., Faul, J. D., Ofstedal, M. B., Langa, K. M., Phillips, J. W., & Weir, D. R. ( 2014). Cohort profile: The health and retirement study (HRS). International Journal of Epidemiology , 43, 576– 585. doi: 10.1093/ije/dyu067 Google Scholar CrossRef Search ADS   Thomas, P. A., Liu, H., & Umberson, D. ( 2017). Family relationships and well-being. Innovation in Aging , 1, 1– 11. doi: 10.1093/geroni/igx025 Ventola, C. L. ( 2014). Social media and health care professionals: Benefits, risks, and best practices. P & T: A Peer-Reviewed Journal For Formulary Management , 39, 491– 520. Walker, J., Crotty, B. H., O’Brien, J., Dierks, M. M., Lipsitz, L., & Safran, C. ( 2017). Addressing the challenges of aging: How elders and their care partners seek information. The Gerontologist , 57, 955– 962. doi: 10.1093/geront/gnw060 Webb, T. L., Joseph, J., Yardley, L., & Michie, S. ( 2010). 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. Journal of Medical Internet Research , 12, e4. doi: 10.2196/jmir.1376 Google Scholar CrossRef Search ADS   Widmer, R. J., Collins, N. M., Collins, C. S., West, C. P., Lerman, L. O., & Lerman, A. ( 2015). Digital health interventions for the prevention of cardiovascular disease: A systematic review and meta-analysis. Mayo Clinic Proceedings , 90, 469– 480. doi: 10.1016/j.mayocp.2014.12.026 Google Scholar CrossRef Search ADS   Xavier, A. J., d’Orsi, E., Wardle, J., Demakakos, P., Smith, S. G., & von Wagner, C. ( 2013). Internet use and cancer-preventive behaviors in older adults: Findings from a longitudinal cohort study. Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology , 22, 2066– 2074. doi: 10.1158/1055-9965.EPI-13-0542 Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Gerontologist Oxford University Press

Internet Use and Preventive Health Behaviors Among Couples in Later Life: Evidence from the Health and Retirement Study

Loading next page...
 
/lp/ou_press/internet-use-and-preventive-health-behaviors-among-couples-in-later-uDzWaEaQKz
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0016-9013
eISSN
1758-5341
D.O.I.
10.1093/geront/gny044
Publisher site
See Article on Publisher Site

Abstract

Abstract Background and Objectives The aim of this study was to examine the link between internet use and preventive health behaviors. We focused on couples to examine whether there were cross-partner associations between internet use and preventive health behaviors. Research Design and Methods The data for this study came from the 2010 and 2012 waves of the Health and Retirement Study and the sample consisted of 5,143 pairs of coupled-individuals. Preventive health behaviors included cancer screenings (mammogram and prostate tests), cholesterol tests, and flu shots. Logistic multilevel actor–partner interdependence models were employed to test the study hypotheses. Results Internet use was associated with a higher likelihood of receiving prostate exams and cholesterol tests for husbands, net of demographic and health characteristics, and insurance status. We found that wives’ internet use was associated with a higher likelihood of receiving flu shots and prostate exams for husbands, but husbands’ internet use was not associated with wives’ preventive health behaviors. Discussion and Implications Research linking internet use and preventive health behaviors is important because such behaviors are associated not only with health of the older population but also with substantial reductions in health care expenditures. Our findings suggested that internet use of older adults is associated with their own preventive health behaviors, as well as their spouses’ preventive health behaviors. Interventions and programs to facilitate older adults’ preventive health behaviors should consider couple-based approaches. Preventive health behaviors, Internet use, Actor–partner interdependence models, Health and Retirement Study Older adults’ use of preventive health care services reduces health risks, affects well-being, and saves lives (Krist et al., 2012). Preventing disease by using health care services not only helps individuals stay healthy, but also has broader positive impacts for the national economy (Centers for Disease Control [CDC], 2017). Researchers have suggested that increasing the use of clinical preventive services could significantly contribute to the health and longevity of older adults in the United States (Farley, Dalal, Mostashari, & Frieden, 2010). For example, cancer screenings such as mammograms and prostate tests when preformed in recommended time periods have been shown to reduce mortality (Centers for Disease Control and Prevention [CDC], Administration on Aging [AOA], Agency for Healthcare Research and Quality [AHRQ], & Centers for Medicare and Medicaid Services [CMS], 2011). Despite the effectiveness of these preventive health services, less than 50% of older Americans aged 65 or older used these services; and even individuals whose insurance provided partial or full coverage for these services received no services or fewer services than for which they were eligible (CDC, 2017; Krist et al., 2012). Specifically, eliminating costs did not seem to significantly increase the use of preventive health care services among older adults (Ozminkowski et al., 2006). To address the chronic disease burden of both individuals and the larger society, various interventions have been implemented and tested (Bauer, Briss, Goodman, & Bowman, 2014). However, broad public support to promote preventive health behaviors and efforts from multiple initiatives have not been highly effective to increase the delivery of preventive health care services (Krist et al., 2012). The literature on the use of preventive health care services of older adults suggested that logistical challenges, lack of knowledge, limited information and motivation, and insufficient collaboration of local and national agencies may work as barriers (Krist et al., 2012; National Prevention Council [NPC], 2011). A common strategy underlying recent efforts to facilitate the use of preventive health care services is providing health-related information using the internet (CDC, AOA, AHRQ, & CMS, 2011; NPC, 2011). Although the link between internet use and access to health information has been examined, prior research has mostly focused on the age related digital health divide (Hall, Bernhardt, Dodd, & Vollrath, 2015; Koch-Weser, Bradshaw, Gualtieri, & Gallagher, 2010; Moorhead et al., 2013), possible harm caused by inaccurate health information on the web (Chesser, Burke, Reyes, & Rohrberg, 2016; Cole, Watkins, & Kleine, 2016), and intervention programs using the internet (Bennett & Glasgow, 2009; Payne, Lister, West, & Bernhardt, 2015; Widmer et al., 2015). Findings from previous studies suggest that the internet can be a source of online health consultations and elimination of false beliefs about certain health conditions (Lo, Esser, & Gordon, 2010; Lu, Shaw, & Gustafson, 2011). However, less is known about of the link between older adults’ internet use and preventive health care behaviors. Further, even less is known about the couple dynamics underlying the link between internet use and preventive health care behaviors (Webb, Joseph, Yardley, & Michie, 2010; Xavier et al., 2013). Most studies to date focused on individual characteristics associated with preventive health care behaviors; this is surprising, given that spouses represent the most important source of care and medical information across the life course (Meyler, Stimpson, & Peek, 2007; Smith, 2011). Findings from earlier studies are also limited by small, non-representative samples and cross-sectional designs (Amante, Hogan, Pagoto, English, & Lapane, 2015; Ayers & Kronenfeld, 2007; Xavier et al., 2013). In this study, we addressed some of these shortcomings by using nationally representative data from the Health and Retirement Study and examining the association between internet use and subsequent preventive health behaviors more than a 2-year study period. Further, we examined whether there are cross-partner associations between internet use and preventive health behaviors. Older Adults’ Internet Use and Preventive Health Behaviors The internet has become an important source of health-related information over the past two decades (Brodie et al., 2000). The expansion of technology has resulted in a rapid increase in the use of the internet to seek health information (Hayward, Hummer, & Sasson, 2015). In addition, a main reason older adults use the internet is to seek health-related information (Heart & Kalderon, 2013). In particular, people who have limited health care services were more likely to use the internet to obtain health information (Amante et al., 2015). Research literature also indicates that internet use influences positive health behaviors such as better diet, daily exercise, and smoking abstinence (Redmond, Baer, Clark, Lipsitz, & Hicks, 2010; Shahab, Brown, Gardner, & Smith, 2014; Xavier et al., 2013). Other studies have shown that the internet provides an accessible resource for obtaining health information (Kivits, 2009), and older adults have also been shown to refer to the internet for health information once they identify a trustworthy source (e.g., WebMD; Walker et al., 2017). Older adults’ acquisition of health information on the internet may influence their health behaviors. As such, health care professionals have been working to develop various web-based interventions and programs to educate older adults with accurate and appropriate health information (Calvillo, Román, & Roa, 2013). Taken together, these findings suggest that internet use may also be associated with other forms of health behaviors, but the link between internet use and preventive health care behaviors remains an understudied area of research. One study showed that older internet users in the United Kingdom were more likely to receive colorectal cancer screenings, but internet use was not associated with receiving mammograms (Xavier et al., 2013). In sum, this literature provides a basis for our hypothesis that internet use would be positively associated with preventive health behaviors among individuals (i.e., actor effects). Older Couples’ Internet Use and Preventive Health Behaviors Life course scholarship has long emphasized the importance of family relationships for well-being in later life. In particular, a substantial body of work has highlighted the principle of linked lives to emphasize the interconnectedness of family relationships and their impact for health and well-being (Elder, Johnson, & Crosnoe, 2003; Gilligan et al., 2017). In this context, the marital dyad is considered to be one of the most important family relationships affecting individuals in later life (Thomas, Liu, & Umberson, 2017). Marital relationships are not only understood to confer significant health benefits, but the health of individuals in marital dyads is also known to be concordant (Meyler et al., 2007; Pai, Godboldo-Brooks, & Edington, 2010). Such health concordance is in part attributable to shared information and health behaviors among couples, as well as the social control function that spouses exert on each other (Thomas et al., 2017). For example, health beliefs and behaviors of a spouse have been shown to influence the other partners’ health behaviors (Manne, Kashy, Weinberg, Boscarino, & Bowen, 2012). Others have shown that marital satisfaction predicts use of cancer screenings among older couples (Kotwal, Lauderdale, Waite, & Dale, 2016). Marriage predicted older men’s colonoscopy use, but not their partners’ use, suggesting some cross-spousal influences in preventive health care behavior (Kotwal et al., 2016). Due to the interdependence of married couples in later life, it is possible that one spouse’s internet use could affect the preventive health behaviors of the other spouse. Therefore, we used a sample of older coupled-individuals to examine cross-partner associations between internet use and preventive health behaviors. Other Factors Affecting Older Adult’s Internet Use and Preventive Health Behaviors Previous literature has demonstrated associations between factors such as race/ethnicity, education, income, and employment status and preventive health care service utilization. For example, older, low-income and ethnic-minority Americans were disproportionately less likely to use these services (CDC, AOA, AHRQ, & CMS, 2011). However, age, education, and race/ethnicity were not significant factors affecting colonoscopy use, whereas higher income was associated with this use (Kotwal et al., 2016). History of medical conditions was also related with preventive health behaviors (Kotwal et al., 2016; Manne et al., 2012). Taken together, this work indicates that certain populations are less likely to use preventive health care services; however, the mechanisms explaining these disparities are somewhat equivocal and not well understood (CDC, AOA, AHRQ, & CMS, 2011; Ozminkowski et al., 2006). Therefore, we will take these demographic factors into consideration when examining the association between older adult’s internet use and preventive health behaviors. In sum, the purpose of this study is to examine the association between internet use and preventive health care behaviors among older couples. Using nationally representative household data for coupled-individuals in the Health and Retirement Study, we investigate whether internet use of older adults and their spouses are prospectively associated with the use of preventive health care services, including influenza vaccinations and cancer screenings, over a 2-year period. Based on the empirical findings discussed previously, we hypothesized that internet use would be positively associated with preventive health care behavior among individuals. We also hypothesized that there would be cross-partner relationships between internet use and preventive health care behavior. Data and Methods Data Source and Study Sample This study was based on the nationally representative Health and Retirement Study (HRS), a biennial longitudinal panel study of individuals over age 51 and their spouses (of any age) in the United States. The HRS, which was initiated in 1992, has collected information from more than 23,000 households (Sonnega et al., 2014). Most of the data for this study came from a file constructed by the Rand Corporation (Version P; Bugliari et al., 2016); measures for internet use came from the publicly available files provided by the Institute of Social Research at the University of Michigan. The key objective of this study was to examine whether internet use was prospectively associated with preventive health care service use. Although the HRS is fielded every 2 years, data on preventive health care are collected every other wave (i.e., every 4 years), with the most recent data available from the 2012 wave. Therefore, we identified the 2010 wave as the baseline point, from which information for the key independent measure (i.e., internet use) and other covariates were assessed (for an exception, see subsequent section on measures for previous preventive health care service utilization); information for preventive health care service utilization was taken from the 2012 wave, which is the follow-up directly following the baseline. The study sample consisted of coupled-individuals who were in a heterosexual marriage (92%) or an otherwise similar partnership (8%). (Same-sex couples were excluded because the analytic technique used in this study is applicable to “distinguishable” dyads only; Kenny & Ledermann, 2010.) The coupled-individuals in the sample were also (a) non-proxy respondents at baseline; and (b) re-interviewed at follow-up. Among couples where both spouses satisfied these criteria (dyad N = 5,188), very few couples had missing information on study variables. That is, 11 couples who had missing information for the internet use measures and 34 couples who had missing information for other study measures were excluded from the sample (0.8%). List-wise deletion was used to handle missing data because there were fewer than 1% missing on any variable in the analysis (cf. Allison, 2010). The final sample included 5,143 coupled-individuals (individual N = 10,286). To ensure that health care service was “preventive” in nature, we applied additional sampling criteria for analyzing specific health care outcomes (Kim & Kawachi, 2017). For example, couples where both spouses did not have a history of stroke or heart disease were selected for the analysis of cholesterol tests (dyad n = 3,132); similarly, couples where both spouses did not have a history of cancer were selected for the analysis of cancer screening (e.g., mammograms and prostate cancer screening; dyad n = 3,958). Of the respective samples, a small fraction of couples had missing information on the outcome variable for one of the spouses (influenza vaccination [dyad n = 12; 0.2%], cholesterol tests [dyad n = 40; 1.3%], and cancer screenings [dyad n = 96; 2.4%]). Because the multilevel analytic approach taken in this study facilitated analysis of couples where a spouse had missing information on the outcome variable (Loeys & Molenberghs, 2013), these couples were retained in the analyses; estimates from sensitivity analyses excluding these couples were consistent with the main findings (results available upon request). Measures Preventive Health Care Behaviors At follow-up, respondents were asked to report on preventive health care services that they received since they were last interviewed (at baseline) with the question: “In the last two years, have you had any of the following medical tests or procedures?” (a) a flu shot, (b) a blood test for cholesterol, (c) a mammogram or x-ray of the breast to search for cancer (for women), and (d) a prostate-specific antigen (PSA) blood test or other examination to screen for prostate cancer (for men). Responses to items regarding flu shots and cholesterol tests were coded dichotomously (1 = yes; 0 = no). A measure for sex-specific cancer screenings was created based on whether the wife and husband had screened for breast and prostate cancer, respectively, which was also coded dichotomously. Information on other preventive health care services were also available from the 2012 wave of the HRS, which included a colonoscopy, sigmoidoscopy, or other screening for colon cancer and a pap smear for cervical cancer. We decided not to include colon cancer screening in our main analysis because 2012 was the first year in which the measure was obtained and no information on previous use was available; an exploratory analysis pertaining to the link between internet use and colon cancer screening is provided in a supplementary material (Supplementary Table 1). Information on Pap smear was also used in a supplementary analysis where female-specific cancer screening was defined as breast cancer or cervical cancer screenings (Supplementary Table 2). Internet Use Respondents’ internet use was assessed at baseline with the following question “Do you regularly use the World Wide Web, or the Internet, for sending and receiving e-mail or for any other purpose, such as making purchases, searching for information, or making travel reservations?” The response to this single-item measure was coded dichotomously (1 = yes; 0 = no). Covariates The HRS contains information on an extensive set of covariates documented or postulated to affect preventive health care behavior, which were considered in the analyses. Demographic characteristics included age (in years), race/ethnic status (non-Hispanic White [reference], non-Hispanic Black, non-Hispanic “other” race, and Hispanic), household income (transformed by the natural log in the analyses), and education (no educational degree [reference], high school diploma or GED, some college, college degree or higher). Health care and health characteristics included insurance coverage through Medicare or Medicaid, either spouse’s employer (current or past), or any other supplemental source (1 = insured; 0 = not insured), whether respondents had seen or talked to medical doctors or nurse practitioners about health over the past 2 years (1 = visited doctor; 0 = no), and number of chronic conditions ever diagnosed by a clinician, including (a) high blood pressure or hypertension, (b) diabetes, (c) cancer (not included for the analyses of cancer screenings), (d) lung disease, (e) heart conditions, (f) stroke, (g) psychiatric problems, and (h) arthritis or rheumatism (range = 0–8). Because preventive health care utilization (i.e., flu shot, cholesterol test, or sex-specific cancer screening) was not assessed at baseline (2010), we created a measure for whether HRS records indicated that respondents had ever received the respective screening or test (1 = used preventive health care; 0 = no). Marriage-related covariates included marital status (1 = married; 0 = non-marital cohabitation), first marriage (1 = first marriage; 0 = others), parental status (1 = has any child(ren); 0 = no child), and length of marriage in years. All covariates except previous preventive health care behavior were measured at baseline. Analytic Strategy The relationship between internet use and subsequent preventive health care utilization during the 2-year observation period was estimated using the multilevel actor–partner interdependence model (APIM) framework widely used to analyze dyadic data (Kenny & Ledermann, 2010). Given that preventive health care service utilization measures were binary, we used generalized estimating equations with an extension for logistic regression models (PROC GEMNOD, SAS Version 9.4; Loeys, Cook, de Smet, Wietzker, & Buysse, 2014). To examine whether internet use was related to preventive health care utilization for both wives and husbands while allowing for the non-independence of the preventive health care utilization within couples, we employed a two-intercept model to estimate the associations. Importantly, the association between one’s own internet use (i.e., actor effects; Hypothesis 1), as well as that of the spouse’s (i.e., partner effects; Hypothesis 2), and preventive health care utilization were simultaneously estimated (Loeys et al., 2014). For each preventive health care behavior, we present two sets of APIMs to test actor and partner effects of internet use. Model 1 included actor and partner effects of internet use without covariate adjustment (i.e., unadjusted model); Model 2 introduced the full set of covariates (i.e., adjusted model). Results Study sample characteristics are presented in Table 1. Approximately 65% of wives and husbands had received an influenza vaccination during the 2-year observation period (gender differences not statistically significant). More than 80% of the sample without a history of a heart disease or stroke at baseline had received a cholesterol test, with wives showing a slightly higher rate compared with husbands. With regard to sex-specific cancer screenings, approximately 72% of wives and 63% of husbands without a history of cancer at baseline had received screenings for breast and prostate cancer, respectively. A majority of the study sample reported using the internet, with wives (62%) showing a higher usage rate compared with husbands (56%). Wives were generally healthier compared with husbands, as indicated by the number of chronic health conditions. Approximately 92% of the couples in the sample were in a formal marriage (as opposed to being in a non-marital partnership). Couples were married for an average of about 30 years. Table 1. Descriptive Characteristics of the Health and Retirement Study Sample   Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —      Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —    Notes: Dyad N = 5,143. M = mean. SD = standard deviation. Group differences were tested using t tests for continuous variables and the chi-square statistics for categorical variables. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). cWhether previous HRS records (1996–2008) indicate preventive health care use. dCouple-level covariates. eInsurance coverage through Medicare/Medicaid, either spouse’s employer (current or past), or any other supplemental insurance. fCount of eight chronic health conditions. gMarital status (1 = married; 0 = non-marital cohabitation). *p < .05. **p < .01. ***p < .001. View Large Table 1. Descriptive Characteristics of the Health and Retirement Study Sample   Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —      Wives  Husbands  t or χ2    M  SD  M  SD  Preventive health behaviors, % (2012)   Flu shot  65.84  —  64.14  —  3.29     Cholesterol testa  83.17  —  80.64  —  6.71  *   Mammogram/X-rayb  71.80  —  —  —  —     Prostate examb  —  —  63.81    —    Previous preventive health behaviors,c %   Flu shot  64.81  —  63.02  —  3.54     Cholesterol testa  86.23  —  84.86    2.34     Mammogram/X-rayb  85.15  —  —    —     Prostate examb  —  —  78.11  —  —    Internet use (2010)  62.41  —  56.02    43.57  ***  Controls (2010)   Age  61.48  (10.65)  64.66  (10.56)  231.61  ***   Race/Ethnicity, %    White (non-Hispanic)  68.46  —  68.44  —  0.60      Black (non-Hispanic)  14.02  —  14.37  —    Other race (non-Hispanic)  3.48  —  3.27  —    Hispanic (any race)  14.04  —  13.92  —   Education, %  1.7762  —  1.774  —        No degree  14.23  —  16.94  —  53.65  ***    High School diploma/GED  35.14  —  31.79  —    Some college  26.95  —  23.24  —    College degree or higher  23.68  —  28.04     Household incomed  10.83  (1.37)            Median value (in $1,000)  56.01             Insured,e %  89.01  —  90.10    3.26     Any doctor visits in 2 years  91.97  —  89.38    20.34  ***   Number of chronic conditionsf  1.75  (1.38)  1.89  (1.46)  24.02  ***   History of cancer, %  11.28  —  13.73    14.10  ***   History of heart disease/stroke, %  17.89  —  27.88    145.42  ***   Marital statusd,g  92.38  —  —    —     First marriage, %  67.11    64.49    7.76  **   Any childrend  95.08  —  —    —     Length of marriage (in years)d  30.38  (17.93)  —    —    Notes: Dyad N = 5,143. M = mean. SD = standard deviation. Group differences were tested using t tests for continuous variables and the chi-square statistics for categorical variables. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). cWhether previous HRS records (1996–2008) indicate preventive health care use. dCouple-level covariates. eInsurance coverage through Medicare/Medicaid, either spouse’s employer (current or past), or any other supplemental insurance. fCount of eight chronic health conditions. gMarital status (1 = married; 0 = non-marital cohabitation). *p < .05. **p < .01. ***p < .001. View Large Influenza Vaccination Table 2 contains results from multilevel models predicting preventive health care use in relation to one’s own and partner’s internet use. The rows containing estimates of actor and partner effects of internet use address Hypotheses 1 and 2, respectively. In the models predicting influenza vaccination, one’s own internet use was not associated with vaccine uptake for either spouse in the unadjusted and adjusted models. Partner’s internet use was also not associated with one’s own vaccine uptake for either spouse in the unadjusted model. However, the partner effect became significant for husbands, but not wives, when covariates were introduced to the model; that is, husbands were more likely to obtain influenza vaccination when their wives used the internet compared with when wives did not use the internet (adjusted OR = 1.22, p < .05). Table 2. Odds Ratios and 95% Confidence Intervals for the Relationship Between Internet Use (2010) and Preventative Health Behaviors (2012)     Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]      Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]  Notes: Models fully adjusted for age, race/ethnic status, education, household income, insurance status, doctor visits, number of chronic health conditions, previous preventive health care utilization, marital status, whether the current marriage is the first marriage, parental status, and length of marriage in years. Dyad N = 5,143; 95% confidence intervals were presented in square brackets. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). *p < .05. **p < .01. ***p < .001. View Large Table 2. Odds Ratios and 95% Confidence Intervals for the Relationship Between Internet Use (2010) and Preventative Health Behaviors (2012)     Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]      Model 1  Model 2  Wife  Husband  Wife  Husband  Flu shot  Actor effects  0.93  1.07  1.03  1.01  [0.81, 1.06]  [0.94, 1.21]  [0.84, 1.25]  [0.83, 1.23]  Partner effects  1.05  0.98  1.05  1.22*  [0.92, 1.19]  [0.86, 1.12]  [0.88, 1.26]  [1.01, 1.47]  Cholesterol testa  Actor effects  1.31*  1.80***  1.16  1.51**  [1.06, 1.63]  [1.48, 2.20]  [0.88, 1.53]  [1.16, 1.97]  Partner effects  1.10  1.33**  0.97  1.11  [0.89, 1.35]  [1.09, 1.63]  [0.75, 1.25]  [0.86, 1.43]  Mammogram/Prostate examb  Actor effects  1.48***  1.69***  1.23  1.38***  [1.26, 1.74]  [1.46, 1.97]  [1.00, 1.51]  [1.14, 1.67]  Partner effects  1.30**  1.34***  1.02  1.20  [1.11, 1.52]  [1.15, 1.56]  [0.85, 1.23]  [1.00, 1.44]  Notes: Models fully adjusted for age, race/ethnic status, education, household income, insurance status, doctor visits, number of chronic health conditions, previous preventive health care utilization, marital status, whether the current marriage is the first marriage, parental status, and length of marriage in years. Dyad N = 5,143; 95% confidence intervals were presented in square brackets. aSubset of the sample where neither spouse has a history of heart disease or stroke (dyad n = 3,132). bSubset of the sample where neither spouse has a history of cancer (dyad n = 3,958). *p < .05. **p < .01. ***p < .001. View Large Cholesterol Test In the unadjusted model, actor effects of internet use were associated with receiving cholesterol tests for both spouses, such that wives (unadjusted OR = 1.31, p < .05) and husbands (unadjusted OR = 1.80, p < .001) who reported using the internet at baseline were more likely to screen for cholesterol compared with those who did not use the internet. In the model fully adjusted for covariates, the actor effect remained significant for husbands (adjusted OR = 1.51, p < .01), but not wives (adjusted OR = 1.16, p = .28). Further, wives’ internet use was significantly associated with husbands’ cholesterol test (unadjusted OR = 1.33, p < .01) in the unadjusted model, but this partner effect was no longer statistically significant in the fully adjusted model (adjusted OR = 1.12, p = .43). Husbands’ internet use was not associated with wives’ cholesterol tests in the unadjusted and adjusted models. Cancer Screening Wives’ and husbands’ internet use was associated with increased likelihood of receiving breast cancer screening for wives (unadjusted OR = 1.48, p < .001) and prostate cancer screening for husbands (unadjusted OR = 1.69, p < .001) in the unadjusted model. The actor effects of internet use remained significant in terms of husbands’ prostate cancer screening in the fully adjusted model (adjusted OR = 1.38, p < .001), but actor effects became marginally significant in the fully adjusted model with regard to wives’ breast cancer screening (adjusted OR = 1.23, p = .052). Also, wives’ internet use was significantly associated with husbands’ prostate cancer screening in the unadjusted (unadjusted OR = 1.34, p < .001), but the effects became marginally significant in the fully adjusted (adjusted OR = 1.20, p = .052) model. Husbands’ internet use was significantly associated with wives’ breast cancer screening in the unadjusted (unadjusted OR = 1.30, p < .01), but the effects were no longer significant in the fully adjusted model. Additional analyses were performed to examine whether the gendered pattern of our findings was statistically significant, using an alternative parameterization of the two-intercept model, known as the “interaction approach” (Loeys et al., 2014). However, results from these additional models indicated that the gender differences in the association between internet use and preventive health care behavior did not reach statistical significance (results available upon request). Discussion Given the disparity in the use of preventive health care services in later life, our central aim in this article was to understand how internet use of older adults and their partners plays a role in preventive health behaviors. Drawing from the life course perspective, which emphasizes the concept of “linked lives,” we examined whether older couples’ internet use was associated with preventive health behaviors over a 2-year period. Specifically, we explored the association between older adults’ internet use and their preventive health behaviors as well as their partners’ internet use in these associations. Overall, the findings from this study indicate that both older adults’ internet use and their partners’ internet use promoted preventive health behaviors. This was true across the range of preventive health behaviors considered, including flu shots, cholesterol tests, mammograms, and prostate exams. We expected that one’s own use of the internet would be positively associated with preventive health behaviors (Hypothesis 1). In regard to cholesterol tests and cancer screenings, our findings supported this hypothesis. In the unadjusted model, internet use was associated with both husbands’ and wives’ receipt of cholesterol and cancer screenings. In the model fully adjusted for covariates (i.e., sociodemographic characteristics, health and health care characteristics, and couple characteristics), effects of husbands’ own internet use remained significant, but the associations between wives’ internet use and preventive health behaviors were no longer statistically significant. Based on the interdependence of couples in later life, we expected cross-partner effects of internet use for both men and women (Hypothesis 2). The same trend we discovered among individuals’ effects of internet use was found between partners. Wives’ internet use was associated with husbands’ receiving flu shots. However, the effects of husbands’ internet use on wives’ preventive health behaviors attenuated when we considered other factors such as education, experience of health care, and couple characteristics. This finding is consistent with previous research, which has demonstrated that women are more knowledgeable on health and health care throughout life course as result of their reproductive health care needs and care-provision responsibilities (Beier & Ackerman, 2003; Chylińska et al., 2017). Based on these needs and roles, women have more active attitudes toward receiving medical treatments (Chylińska et al., 2017). In addition, women are more likely than men to have early and frequent exposures to health care services (Kent et al., 2017). As a result, some health-related information from the internet may be less relevant to women’s own preventive health behaviors. In line with previous research, which has demonstrated that married people influence their partners’ health in later-life (Meyler et al., 2007; Pai et al., 2010), our findings indicate that older adults’ internet use is associated with their own preventive health behaviors, as well as their spouses’ preventive health behaviors. However, given that the gender differences in the association between wives and husbands in the sample did not reach statistical significance, more research is needed to corroborate the gendered pattern of findings reported in this study. Future Directions The internet use measure included in this study assessed the general use of the internet and is not specific to information acquisition in relation to preventive health behaviors. Thus, the potential pathways linking internet use and preventive health behaviors were not directly evaluated because this information was not available in the Health and Retirement Study. Future research should consider these associations between health-related internet use and preventative health behaviors. Second, a strength of this study was its ability to examine the associations between internet use and preventive health behaviors among older couples. However, recent research indicates the growing prevalence of alternative union forms (Benson & Coleman, 2016; Fredriksen-Goldsen, Hoy-Ellis, Goldsen, Emlet, & Hooyman, 2014). Future research should consider whether older adults in these alternative union forms also receive health benefits from internet use. Further, because the study focused on couples, our findings are not generalizable to older individuals who are not in relationships. Given that marriage has health benefits for older adults (Meyler et al., 2007; Pai et al., 2010), single older adults are particularly vulnerable to health problems. Implications for Practice Taken together, the findings in this study support the documented positive impacts of older adults’ internet use on their health behaviors by specifically highlighting the role internet use has in older adult use of preventative health care services (Bauer et al., 2014; Cheston, Flickinger, & Chisolm, 2013; Ventola, 2014). Research suggests that internet use is not only a source of health-related information, but it can also be an effective intervention strategy to promote health among older adults (Cheston et al., 2013). Specifically, professionals working on the development of interventions and programs to promote preventive health behaviors have highlighted the interactive features of the internet, and web-based programs are expanding rapidly (CDC, 2011; Cheston et al., 2013; Hall et al., 2015). Thus, there is growing interest in developing programs and interventions to promote older adults’ health behaviors. The findings from this study indicated that internet use may be a particularly promising approach to increase older adults’ use of preventative health care services and should be considered in future intervention efforts. Further, given our findings indicating that older adults benefit not just from their own internet use but also their spouses, it may be particularly advantageous to include both partners in programming efforts. Current efforts to increase older adults’ use of preventive health care services have focused almost exclusively on individuals (Arden-Close & McGrath, 2017). Our findings suggest that programs that incorporate dyadic designs may be particularly promising to facilitate older adults’ internet use and engagement in preventative health behaviors (Arden-Close & McGrath, 2017). Supplementary Data Supplementary data are available at The Gerontologist online. References Allison, P. ( 2010). Missing data. In J. Wright & P. Marsden (Eds.), Handbook of survey research  (pp. 631– 658). Bingley, UK: Emerald Group Limited. Amante, D. J., Hogan, T. P., Pagoto, S. L., English, T. M., & Lapane, K. L. ( 2015). Access to care and use of the Internet to search for health information: Results from the US National Health Interview Survey. Journal of Medical Internet Research , 17, e106. doi: 10.2196/jmir.4126 Google Scholar CrossRef Search ADS   Arden-Close, E., & McGrath, N. ( 2017). Health behaviour change interventions for couples: A systematic review. British Journal of Health Psychology , 22, 215– 237. doi: 10.1111/bjhp.12227 Google Scholar CrossRef Search ADS   Ayers, S. L., & Kronenfeld, J. J. ( 2007). Chronic illness and health-seeking information on the Internet. Health (London, England: 1997) , 11, 327– 347. doi: 10.1177/1363459307077547 Google Scholar CrossRef Search ADS   Bauer, U. E., Briss, P. A., Goodman, R. A., & Bowman, B. A. ( 2014). Prevention of chronic disease in the 21st century: Elimination of the leading preventable causes of premature death and disability in the USA. Lancet (London, England) , 384, 45– 52. doi: 10.1016/S0140-6736(14)60648-6 Google Scholar CrossRef Search ADS   Beier, M. E., & Ackerman, P. L. ( 2003). Determinants of health knowledge: An investigation of age, gender, abilities, personality, and interests. Journal of Personality and Social Psychology , 84, 439– 448. doi:10.1037/0022-3514.84.2.439 Google Scholar CrossRef Search ADS   Bennett, G. G., & Glasgow, R. E. ( 2009). The delivery of public health interventions via the Internet: Actualizing their potential. Annual Review of Public Health , 30, 273– 292. doi: 10.1146/annurev.publhealth.031308.100235 Google Scholar CrossRef Search ADS   Benson, J. J., & Coleman, M. ( 2016). Older adults developing a preference for living apart together. Journal of Marriage and Family , 78, 797– 812. doi: 10.1111/jomf.12292 Google Scholar CrossRef Search ADS   Brodie, M., Flournoy, R. E., Altman, D. E., Blendon, R. J., Benson, J. M., & Rosenbaum, M. D. ( 2000). Health information, the Internet, and the digital divide. Health affairs (Project Hope) , 19, 255– 265. doi:10.1377/hlthaff.19.6.255 Google Scholar CrossRef Search ADS   Bugliari, D., Campbell, N., Chan, C., Hayden, O., Hurd, M., Main, R., … St. Clair, P. ( 2016). Rand HRS data documentation, version p . Santa Monica, CA: RAND Center for the Study of Aging, with funding from the National Institute on Aging and the Social Security Administration. Calvillo, J., Román, I., & Roa, L. M. ( 2013). Empowering citizens with access control mechanisms to their personal health resources. International Journal of Medical Informatics , 82, 58– 72. doi: 10.1016/j.ijmedinf.2012.02.006 Google Scholar CrossRef Search ADS   Centers for Disease Control.( 2017). Preventive health care . Retrieved from https://www.cdc.gov/healthcommunication/toolstemplates/entertainmented/tips/PreventiveHealth.html Centers for Disease Control and Prevention, Administration on Aging, Agency for Healthcare Research and Quality, & Centers for Medicare and Medicaid Services. ( 2011). Enhancing use of clinical preventive services among older adults . Washington, DC: AARP. PubMed PubMed  Chesser, A., Burke, A., Reyes, J., & Rohrberg, T. ( 2016). Navigating the digital divide: A systematic review of eHealth literacy in underserved populations in the United States. Informatics for Health & Social Care , 41, 1– 19. doi: 10.3109/17538157.2014.948171 Google Scholar CrossRef Search ADS   Cheston, C. C., Flickinger, T. E., & Chisolm, M. S. ( 2013). Social media use in medical education: A systematic review. Academic Medicine: Journal of the Association of American Medical Colleges , 88, 893– 901. doi: 10.1097/ACM.0b013e31828ffc23 Google Scholar CrossRef Search ADS   Chylińska, J., Łazarewicz, M., Rzadkiewicz, M., Adamus, M., Jaworski, M., Haugan, G., … Włodarczyk, D. ( 2017). The role of gender in the active attitude toward treatment and health among older patients in primary health care–self-assessed health status and sociodemographic factors as moderators. BMC Geriatrics , 17, 284. doi: 10.1186/s12877-017-0677-z Google Scholar CrossRef Search ADS   Cole, J., Watkins, C., & Kleine, D. ( 2016). Health advice from internet discussion forums: How bad is dangerous? Journal of Medical Internet Research , 18, e4. doi: 10.2196/jmir.5051 Google Scholar CrossRef Search ADS   Elder, G. H. Jr., Johnson, M. K., & Crosnoe, R. ( 2003). The emergence and development of life course theory. In J. T. Mortimer & M. J. Shanahan (Eds.), Handbook of the life course  (pp. 3– 19). New York, NY: Springer. Google Scholar CrossRef Search ADS   Farley, T. A., Dalal, M. A., Mostashari, F., & Frieden, T. R. ( 2010). Deaths preventable in the U.S. by improvements in use of clinical preventive services. American Journal of Preventive Medicine , 38, 600– 609. doi: 10.1016/j.amepre.2010.02.016 Google Scholar CrossRef Search ADS   Fredriksen-Goldsen, K. I., Hoy-Ellis, C. P., Goldsen, J., Emlet, C. A., & Hooyman, N. R. ( 2014). Creating a vision for the future: Key competencies and strategies for culturally competent practice with lesbian, gay, bisexual, and transgender (LGBT) older adults in the health and human services. Journal of Gerontological Social Work , 57( 2–4), 80– 107. doi: 10.1080/01634372.2014.890690 Google Scholar CrossRef Search ADS   Gilligan, M., J. J. Suitor, Nam, S., Routh, B., Rurka, M., & Con, G. ( 2017). Family networks and psychological well-being in midlife. Social Sciences , 6, 94. doi: 10.3390/socsci6030094 Google Scholar CrossRef Search ADS   Hall, A. K., Bernhardt, J. M., Dodd, V., & Vollrath, M. W. ( 2015). The digital health divide: Evaluating online health information access and use among older adults. Health Education & Behavior: The Official Publication of the Society for Public Health Education , 42, 202– 209. doi: 10.1177/1090198114547815 Google Scholar CrossRef Search ADS   Hayward, M. D., Hummer, R. A., & Sasson, I. ( 2015). Trends and group differences in the association between educational attainment and U.S. adult mortality: Implications for understanding education’s causal influence. Social Science & Medicine (1982) , 127, 8– 18. doi: 10.1016/j.socscimed.2014.11.024 Google Scholar CrossRef Search ADS   Heart, T., & Kalderon, E. ( 2013). Older adults: Are they ready to adopt health-related ICT? International Journal of Medical Informatics , 82, e209– e231. doi: 10.1016/j.ijmedinf.2011.03.002 Google Scholar CrossRef Search ADS   Kenny, D. A., & Ledermann, T. ( 2010). Detecting, measuring, and testing dyadic patterns in the actor-partner interdependence model. Journal of Family Psychology: JFP , 24, 359– 366. doi: 10.1037/a0019651 Google Scholar CrossRef Search ADS   Kent, S., Fusco, F., Gray, A., Jebb, S. A., Cairns, B. J., & Mihaylova, B. ( 2017). Body mass index and healthcare costs: A systematic literature review of individual participant data studies. Obesity Reviews: An Official Journal of the International Association for the Study of Obesity , 18, 869– 879. doi: 10.1111/obr.12560 Google Scholar CrossRef Search ADS   Kim, E. S., & Kawachi, I. ( 2017). Perceived neighborhood social cohesion and preventive healthcare use. American Journal of Preventive Medicine , 53, e35– e40. doi: 10.1016/j.amepre.2017.01.007 Google Scholar CrossRef Search ADS   Kivits, J. ( 2009). Everyday health and the internet: A mediated health perspective on health information seeking. Sociology of Health & Illness , 31, 673– 687. doi: 10.1111/j.1467-9566.2008.01153.x Google Scholar CrossRef Search ADS   Koch-Weser, S., Bradshaw, Y. S., Gualtieri, L., & Gallagher, S. S. ( 2010). The Internet as a health information source: Findings from the 2007 Health Information National Trends Survey and implications for health communication. Journal of Health Communication , 15( Suppl. 3), 279– 293. doi: 10.1080/10810730.2010.522700 Google Scholar CrossRef Search ADS   Kotwal, A. A., Lauderdale, D. S., Waite, L. J., & Dale, W. ( 2016). Differences between husbands and wives in colonoscopy use: Results from a national sample of married couples. Preventive Medicine , 88, 46– 52. doi: 10.1016/j.ypmed.2016.03.011 Google Scholar CrossRef Search ADS   Krist, A., Shenson, D., Woolf, S., Bradley, C., Liaw, W., & Rothemich, S. ( 2012). A framework for integration of community and clinical care to improve the delivery of clinical preventive services among older adults . National Association of Chronic Disease Directors and the Michigan Public Health Institute. Retrieved from https://www.cdc.gov/aging/pdf/cps-integration-framework.pdf Lo, A. S., Esser, M. J., & Gordon, K. E. ( 2010). YouTube: A gauge of public perception and awareness surrounding epilepsy. Epilepsy & Behavior: E&B , 17, 541– 545. doi: 10.1016/j.yebeh.2010.02.004 Google Scholar CrossRef Search ADS   Loeys, T. O. M., Cook, W., de Smet, O., Wietzker, A., & Buysse, A. N. N. ( 2014). The actor-partner interdependence model for categorical dyadic data: A user-friendly guide to GEE. Personal Relationships , 21, 225– 241. doi: 10.1111/pere.12028 Google Scholar CrossRef Search ADS   Loeys, T., & Molenberghs, G. ( 2013). Modeling actor and partner effects in dyadic data when outcomes are categorical. Psychological Methods , 18, 220– 236. doi: 10.1037/a0030640 Google Scholar CrossRef Search ADS   Lu, H. Y., Shaw, B. R., & Gustafson, D. H. ( 2011). Online health consultation: Examining uses of an interactive cancer communication tool by low-income women with breast cancer. International Journal of Medical Informatics , 80, 518– 528. doi: 10.1016/j.ijmedinf.2011.03.011 Google Scholar CrossRef Search ADS   Manne, S., Kashy, D., Weinberg, D. S., Boscarino, J. A., & Bowen, D. J. ( 2012). Using the interdependence model to understand spousal influence on colorectal cancer screening intentions: A structural equation model. Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine , 43, 320– 329. doi: 10.1007/s12160-012-9344-y Google Scholar CrossRef Search ADS   Meyler, D., Stimpson, J. P., & Peek, M. K. ( 2007). Health concordance within couples: A systematic review. Social Science & Medicine (1982) , 64, 2297– 2310. doi: 10.1016/j.socscimed.2007.02.007 Google Scholar CrossRef Search ADS   Moorhead, S. A., Hazlett, D. E., Harrison, L., Carroll, J. K., Irwin, A., & Hoving, C. ( 2013). A new dimension of health care: Systematic review of the uses, benefits, and limitations of social media for health communication. Journal of Medical Internet Research , 15, e85. doi: 10.2196/jmir.1933 Google Scholar CrossRef Search ADS   National Prevention Council.( 2011). National prevention strategy . Washington, DC: U.S. Department of Health and Human Services, Office of the Surgeon General. Ozminkowski, R. J., Goetzel, R. Z., Shechter, D., Stapleton, D. C., Baser, O., & Lapin, P. ( 2006). Predictors of preventive service use among Medicare beneficiaries. Health Care Financing Review , 27, 5– 23. Pai, C. W., Godboldo-Brooks, A., & Edington, D. W. ( 2010). Spousal concordance for overall health risk status and preventive service compliance. Annals of Epidemiology , 20, 539– 546. doi: 10.1016/j.annepidem.2010.03.020 Google Scholar CrossRef Search ADS   Payne, H. E., Lister, C., West, J. H., & Bernhardt, J. M. ( 2015). Behavioral functionality of mobile apps in health interventions: A systematic review of the literature. JMIR Mhealth and Uhealth , 3, e20. doi: 10.2196/mhealth.3335 Google Scholar CrossRef Search ADS   Redmond, N., Baer, H. J., Clark, C. R., Lipsitz, S., & Hicks, L. S. ( 2010). Sources of health information related to preventive health behaviors in a national study. American Journal of Preventive Medicine , 38, 620– 627.e2. doi: 10.1016/j.amepre.2010.03.001 Google Scholar CrossRef Search ADS   Shahab, L., Brown, J., Gardner, B., & Smith, S. G. ( 2014). 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. Journal of Medical Internet Research , 16, e253. doi: 10.2196/jmir.3368 Google Scholar CrossRef Search ADS   Smith, D. ( 2011). Health care consumer’s use and trust of health information sources. Journal of Communication in Healthcare , 4, 200– 210. doi: 10.1179/1753807611Y.0000000010 Google Scholar CrossRef Search ADS   Sonnega, A., Faul, J. D., Ofstedal, M. B., Langa, K. M., Phillips, J. W., & Weir, D. R. ( 2014). Cohort profile: The health and retirement study (HRS). International Journal of Epidemiology , 43, 576– 585. doi: 10.1093/ije/dyu067 Google Scholar CrossRef Search ADS   Thomas, P. A., Liu, H., & Umberson, D. ( 2017). Family relationships and well-being. Innovation in Aging , 1, 1– 11. doi: 10.1093/geroni/igx025 Ventola, C. L. ( 2014). Social media and health care professionals: Benefits, risks, and best practices. P & T: A Peer-Reviewed Journal For Formulary Management , 39, 491– 520. Walker, J., Crotty, B. H., O’Brien, J., Dierks, M. M., Lipsitz, L., & Safran, C. ( 2017). Addressing the challenges of aging: How elders and their care partners seek information. The Gerontologist , 57, 955– 962. doi: 10.1093/geront/gnw060 Webb, T. L., Joseph, J., Yardley, L., & Michie, S. ( 2010). 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. Journal of Medical Internet Research , 12, e4. doi: 10.2196/jmir.1376 Google Scholar CrossRef Search ADS   Widmer, R. J., Collins, N. M., Collins, C. S., West, C. P., Lerman, L. O., & Lerman, A. ( 2015). Digital health interventions for the prevention of cardiovascular disease: A systematic review and meta-analysis. Mayo Clinic Proceedings , 90, 469– 480. doi: 10.1016/j.mayocp.2014.12.026 Google Scholar CrossRef Search ADS   Xavier, A. J., d’Orsi, E., Wardle, J., Demakakos, P., Smith, S. G., & von Wagner, C. ( 2013). Internet use and cancer-preventive behaviors in older adults: Findings from a longitudinal cohort study. Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology , 22, 2066– 2074. doi: 10.1158/1055-9965.EPI-13-0542 Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The GerontologistOxford University Press

Published: May 22, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off