This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Aging, is properly cited. The complete bibliographic information, a link to the original publication on http://aging.jmir.org, as well as this copyright and license information must be included.
As individuals age, chronic health difficulties may disrupt physical and social well-being. Individuals can turn to online communities to interact with similar peers, which may help buffer negative effects resulting from health difficulties.
This study investigated the reasons that older adults join a diabetes online community to better understand the specific resources that are being sought.
We used semantic network analyses to categorize the reasons participants provided for joining a community during the sign-up process.
The most frequent reasons for joining were to seek information about their health condition, to help with self-management of health difficulties, for feelings of informational and social support, and for having a community with whom to share. Women were more likely to go online for sharing and companionship as well as for information and social support reasons, whereas men were more likely to go online for general information and self-management reasons.
This study shows the reasons older adults seek to join a diabetes online community: for increased information and support regarding chronic health difficulties. Practitioners may want to consider ways to promote access to online communities among their older patients as a source of health information and a resource to provide a sense of community.
Across one’s lifespan, social interactions with same-aged peers—who are more likely to share attitudes, values, and interests—are important [
In this study, we examined the reasons that older adults provided for joining an online community during their initial registration, to shed light on their needs and goals. The socioemotional selectivity theory suggests that older adults, due to perceived limitations on time and energy, are more likely to invest time
Although older adults receive health information from their primary care providers, seeking supplementary health information is still one of the most popular online activities [
Moreover, an online environment allows older adults to send and receive information to and from others asynchronously, thus reducing any restriction on time and mobility for receiving information about their condition [
More specifically, however, older adults likely join online communities to seek information related to self-management of a chronic health condition, which has been shown to contribute to older adults’ quality of life [
In addition to being a valuable resource for seeking health information, social support for chronic health difficulties may be another reason for joining online communities [
Social support is especially valued when it comes from individuals with similar experiences [
In this study, we investigated three general areas that older adults may offer as reasons they joined an online health community: (1) health information seeking, (2) self-regulation, and (3) social support. We drew on data from one of the largest diabetes online communities in the United States, the Diabetes Hands Foundation (DHF). The DHF was a nonprofit organization that “connects, engages and empowers people touched by diabetes.” At this time, the DHF has resolved and TuDiabetes is now part of the
The dataset included limited demographic information including age, sex, and diabetes type (I or II). The reason for joining was obtained from an open-ended question, “Why did you want to join?” Data for this study were obtained between June 12, 2007, and September 1, 2014, after which TuDiabetes began using a new database and no longer asked this question on joining. The dataset was retrieved in December 2014. Permission for this study was obtained from the Institutional Review Board at Northwestern University.
Inclusion criteria for the study were that members had to be at least 65 years old and have type II diabetes. Age was reported by members at sign-up. The database contained 34,797 records: 30,248 participants were younger than 65 years old, 435 had type I diabetes, 49 had pre/no diabetes, and 2096 did not specify their age. The final sample included 1969 individuals, aged 65 and over, with type II diabetes.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Consent was obtained from DHF for all participants included in the study.
We analyzed the unstructured free-text field responses that members provided on joining, in two phases.
First, we examined the content of each response using semantic network analysis, which assesses the frequency of word co-occurrences [
The output of the network analysis can be seen in
Borrowing the approach used by Wang et al [
general information if they indicated that the new member sought advice, referrals, or knowledge [
self-management if the word pairs indicated older adults going to the DHF for help with activities such as diet, self-regulation, pump, or medicine [
share/support/companionship if the word pairs indicated anything involving two or more people and did not include words such as support, help, or advice [
informational support if the word pairs were informational in nature and included words such as support, help, or advice without mentioning another person [
social support if the word pairs were social in nature and included words such as support, help, or advice while mentioning another person [
We coded the top 100 word pairs (a weight of 7077/26,685) to determine their relative frequency. Word pairs were coded independently by 2 raters, yielding adequate reliability (Cohen’s kappa = .73). A third rater resolved disagreements.
In sum, we conducted two sets of analyses on these words used by older adults: item-level and person-level. Item-level analyses were conducted to assess the frequency of word co-occurrences. Person-level analyses were conducted to examine possible individuals’ differences in reasons for joining the DHF. Word pairs were always coded into the most specific categories if possible (self-management, share/support/companionship, informational support, social support). If word pairs could not be coded into the specific categories but were informational in nature, we coded them as general information. Less than 2% of the word pairs could not be coded into a category.
Degree of centrality of the words. This figure illustrates the relationship among the top 100 pairs: the more centered the words, the more significant they are.
The strength of relationship between the top 15 word-pairs (the edges): the thicker the lines, the stronger the relationship for those word-pairs.
Definitions of the coded categories and word pair examples.
Category | Definition | Actual word pair examples | References |
Information | knowledge related | diabetes information; disease information; learn how | Bartlett & Coulson, 2011; Greene et al, 2010; Kaufman, 2010 |
Support (informational) | “support,” “help,” “advice,” without referencing another person | support how; information help | Menefee et al, 2016; Kaufman, 2010 |
Support (social) | “support,” “help,” “advice,” referencing another person | support people; help people | Crotty et al, 2015; Dale et al, 2012; Nicklett et al, 2013 |
Share/companionship | “share,” and involve other people | information community; share support; friendship information | Crotty et al, 2015; Strom & Egede, 2012; Vassilev et al, 2013 |
Self-management | diet, self-regulation, complication, blood, etc; pump and instrument; medicine | control better; treatment information; recipes information; pump information; diabetes pump; understand Super Bolus; insulin questions | Bodenheimer et al, 2002; Quinn et al, 2011; Vassilev et al, 2013 |
N/A | not applicable | other more; other how |
Item analyses (
Person-level results are shown in
In addition to examining the percentage of older adults that endorsed the five categories or reasons for going online and interacting within the DHF, we also were interested in exploring whether older adult men and women in this sample differ in the rates that they endorse their respective reasons for joining the DHF. A chi-square test was run to determine whether men and women in this sample endorsed the reasons for joining the DHF at similar rates. Older adult men and women did not endorse each of the five reasons for joining the DHF at the same rates: χ24=16.172 (N=1559),
Categories of word pairs (N=7077).
Category | n (%) |
General information | 3223 (45.54) |
Other/Uncategorized | 141 (1.99) |
Self-management | 1191 (16.83) |
Share/Support/Companionship | 981 (13.86) |
Support (informational) | 1121 (15.84) |
Support (social) | 420 (5.94) |
Reasons members gave for joining TuDiabetes (N=1969).
Category | n (%) |
Health information seeking | 575 (29.20) |
Other/Uncategorized | 408 (20.72) |
Self-management | 364 (18.50) |
Share/Support/Companionship | 110 (5.60) |
Support (informational) | 356 (18.10) |
Support (social) | 156 (7.92) |
Cross-tabulation of gender by category (N=1559).
Category | Gendera, n (%) | ||
Men (n=789) | Women (n=770) | ||
General information | 323 (41.94) | 251 (32.59) | |
Self-management | 186 (23.57) | 179 (23.25) | |
Share/support/companionship | 168 (21.29) | 188 (24.42) | |
Information support | 64 (8.11) | 89 (11.56) | |
Social support | 48 (6.08) | 63 (8.18) |
aχ2=16.172 (df=4);
Even with barriers to social contact in older age such as limitations on mobility as the result of health difficulties, online communities may be one way for individuals to have social contact regardless of time, location, or physical ability [
The results from this study add to the literature in a several ways. First, previous studies have shown that older adults do use online communities to obtain health information related to chronic conditions [
The results of this study do not appear to fully support the socioemotional selectivity theory [
In general, the data show that both information and social support are key reasons why older adults join online health communities. More work is needed to examine the interactions between obtaining and using health information on the one hand and feeling socially connected to similar peers on the other. Past work has shown that high levels of engagement in diabetes online communities is associated with better glycemic levels, diabetes self-care, and health-related quality of life [
There are several limitations to our study. First, we relied on a naturalistic dataset with an open-ended question on reasons for joining that was likely interpreted differently across individuals and that did not provide an opportunity for follow-up questions when general responses were provided. In addition, we did not obtain information on continued use of the online community or on community members’ income or education levels, both of which are related to online use [
Additionally, we must consider the data reported here in light of the growth of social media use in recent years, that participants could have increasing alternatives for online communities. However, according to the Pew Research Center, Twitter use today remains very low among older adults (8%). While Facebook use is higher (41%), the majority of older adults do not use it for a specific purpose [
Our findings suggest that older adults seek online communities for specific types of information regarding their chronic health conditions. As such, when designing an online community for use by older adults, it should be created so that it is easy for individuals to seek information from and share information with similar others, especially as it relates to medications and other self-management practices (technology tools). In addition, the results show that older adults seek online communities for social support. While older adults may be given sufficient health information from their primary care provider, they may find it useful to connect with similar others to better understand the information and how to apply it to their condition [
Diabetes Hands Foundation
We would like to thank the Diabetes Hands Foundation for sharing their data with our research team. This work was supported by the National Science Foundation (NSF GRFP 1650042).
JAL participated in interpreting the analyses and drafted the initial manuscript and revisions. PMG conceived of the initial idea for the study, gained access to a dataset, participated in performing and interpreting the analyses, and helped develop the manuscript. CLH participated in performing and interpreting the analyses and helped draft the Methods and Results sections. LMSM conceived of the initial idea for the study, participated in performing and interpreting the analyses, and helped develop the manuscript and subsequent revisions. All authors read and approved the final copy.
None declared.