Online social support for obese adults: Exploring the role of forum activity

International Journal of Medical Informatics 101 (2017) 1–8

Contents lists available at ScienceDirect

International Journal of Medical Informatics journal homepage: www.ijmijournal.com

Online social support for obese adults: Exploring the role of forum activity Doreen Reifegerste a,∗ , Katrin Wasgien b , Lutz M. Hagen b a b

Department of Journalism and Communication Research, Hanover University of Music, Drama and Media, Germany Institute of Media and Communication, University of Technology Dresden, Germany

a r t i c l e

i n f o

Article history: Received 24 September 2015 Received in revised form 18 January 2017 Accepted 3 February 2017 Keywords: Obesity Social support Online forum Active participation

a b s t r a c t Purpose: Worldwide, the number of obese persons continues to grow. Online-mediated self-help groups represent an opportunity for obese persons to support each other. The aim of our study was to evaluate whether and how the use of and active participation in online self-help groups is associated with perceived informational and emotional support among obese adults. Methods: We conducted an online-based questionnaire (N = 230) with users of online self-help groups for obese adults in Germany. Results: Findings revealed that forum activity is significantly correlated with perceived informational and emotional support. While asking questions was strongly correlated with both types of social support, sharing opinions and answering posts were more strongly correlated with perceived emotional support. Conclusion: The level of social support in online communities depends on an individual’s forum activity. Our findings offer a foundation for professionals in the health care sector to enhance their understanding, make recommendations, and further develop online self-help groups. © 2017 Elsevier B.V. All rights reserved.

1. Introduction In the last decade online support groups have become an increasingly common way to obtain support for health-related issues [1,2]. Online support groups address a range of health issues such as cancer, HIV, and pregnancy, and can help people to receive relevant and specific information, as well as to effectively manage stress and cope with negative events resulting from their health problems. Online communication enables the asynchronous and anonymous exchange of social support [3,4], which engages social group processes for remote participants and disinhibits communication among weak-tie networks [5]. This offers the possibility of social support to populations that would not otherwise be able to obtain it. Online support groups enable members of a group to support each other despite certain impediments (such as mobility), while also representing a low-cost supplement to the traditional professional treatment of obesity [5]. The fact that physical appearance is not evaluated in online support groups [6] may also be an important consideration for obese persons, who are often confronted with stigmatization and discrimination [7]. The benefits of online obesity

∗ Corresponding authors. E-mail address: [email protected] (D. Reifegerste). http://dx.doi.org/10.1016/j.ijmedinf.2017.02.003 1386-5056/© 2017 Elsevier B.V. All rights reserved.

support groups may be significant, as this disease has dramatically increased worldwide in recent decades [8], posing serious health risks and increasing related medical costs [9,10]. To date, studies evaluating the efficacy of online interventions for obese persons have focused on participant weight loss as the key outcome measure [11]. However, the treatment of obesity should not be limited to weight loss, as sustainable healthy behaviors are important for sustaining success. In addion, online groups are—in contrast to clinical therapies—more focused on improving general psychological outcomes than on specific interventions aims [5] or health parameters [3]. Online groups should therefore not be regarded as a substitution for weight management programs, but rather as a complement to them. Das and Faxvaag [12] regard online forums as particularly useful for those patients who exclude themselves from traditional interventions or who experience barriers to making contact with medical experts. Therefore, participation in online support networks should be considered an important part of obesity treatment programs. The present paper presents current research on obesity therapy and theories that can be applied to social support interventions in obesity therapy. Our study explores the role that online social support plays for obese or overweight individuals. We conducted an online survey to investigate the associations between online forum activity and different types of perceived social support. The present paper concludes with a discussion of the implications of how online

2

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

support networks can be used as part of obesity treatment programs with regard to health communication theory; it also presents ideas for future research. 2. Challenges of obesity According to the WHO [10], being overweight or obese is defined as having abnormal or excessive fat accumulation in relation to one’s weight and height (i.e.; Body Mass Index), to a degree that is hazardous to one’s health. While a Body Mass Index (BMI) lower than 25 is classified as normal weight, individuals with a higher BMI are classified as overweight (≥25) or obese (BMI ≥ 30) by the WHO [10]. About 13% of the world’s adult population, 23% of adults in Germany, and more than 33% of adults in the United States are considered obese, and more than 39% of adults worldwide are overweight [13,14,10]. The prevalence of being overweight or obese was once considered a problem only in high income countries; however, this worrisome health trend is now dramatically on the rise in lowand middle-income countries as well [10]. Being obese is currently regarded as a more serious health risk than being underweight [10], with obesity identified as a major risk factor for a number of chronic diseases, including diabetes, cardiovascular diseases, and cancer [14]. In Germany, obesity is among the most common underlying causes of preventable diseases and deaths [59]. Being obese or overweight also often causes considerable limitations in the social life, mobility, and quality of life of affected individuals [13]. Obese persons face a number of barriers; some of these barriers are based on having difficulty in moving or going outside, while others are more social, and based on shame or social exclusion [15,16]. The treatment of obesity is connected with various challenges. In addition to traditional medical treatments (i.e., pharmaceutical and surgical interventions) of physical symptoms and comorbidities, lifestyle modifications (including diet, exercise, and behavior therapy) are common tools in obesity treatment interventions [17]. Obesity is a chronic disease [9] that cannot be cured by any single measure, and treatment is lengthy and includes the threat of relapses [17]; any successful treatment must therefore have long-term effects [18]. To maintain behavioral changes and the associated weight loss, individuals must develop self-management [19]; individual success in building this core competence is in turn based on social, psychological, and competence-related variables that are often not addressed in programs that concentrate solely on weight loss, or that are primarily commercial [20]. In addition, the widespread stigmatization of obese individuals often contributes to low self-esteem and feelings of low self-control; the resulting reduced feelings of self-efficacy can lead to difficulties in developing the constructive coping strategies necessary for weight loss [7]. With these broader causes and challenges in mind, the focus of behavior therapy is not on weight loss per se, but on the empowerment, self-management, and self-regulation [21] necessary to achieve this change. These personal strengths can be nurtured by both offline and online disease-specific social support groups, which offer peer counseling and various forms of social support [9]. 3. Theories of social support Bandura [22] views social support as an important antecedent for both self-efficacy beliefs and behavior change. According to his social cognitive theory, individuals who perceive themselves to have stronger social support are likely to hold stronger selfefficacy beliefs, which positively influence behavior changes and the maintenance of those changes. Several studies have demonstrated the relevance of informational and emotional social support

[23,12,24,25] in the efficacy of both online and offline obesity interventions. These studies indicate that social support is important for long-term weight loss maintenance. As social support is such an important prerequisite in the treatment of obesity [26], it should therefore be considered an important factor in the success of online support groups for obese persons [27]. Harvey-Berino et al. [18] could already show that internet is a viable medium for promoting long-term weight loss in a 12-month support intervention program. Many studies have examined the social support provided via online support groups for various health-related problems and topics, and these studies have provided ample evidence of the promising potential of these forums (for a review, see [28,29]. Online support groups can augment the social support of one’s offline network [30], increase the perceived availability of social support [6], provide very specific informational support [29], and foster compliance [31] and empowerment by enabling emotional relief [5]. Although social support includes many dimensions, online networks most frequently feature informational and emotional support [32–35]. Informational support pertains to the exchange of relevant information and advice, while emotional support is the provision of caring and sympathy. Content analysis revealed that both forms of social support are highly relevant in online support groups, with different health topics focusing on different forms. For example, forums for people with disabilities may focus on emotional support [36], while informational support messages may be more frequent for individuals living with irritable bowel syndrome [37] or AIDS [34]. The theory of optimal matching [60] states that, depending on the kind of stress, different forms of social support may be most beneficial. For example, while controllable events benefit most from informational support, uncontrollable events require more emotional support. Turner et al. [30] argue that online communities can dramatically increase opportunities for optimal matching to occur, as, unlike in scattered offline networks, thousands of other participants with similar concerns are available. This argument is also supported by weak-tie support network theory [2], which considers online communities to be one way to provide access to weak-tie support. This effect is especially important in cases where strong-tie support may be perceived to be inadequate or insufficient (e.g., because close friends or family members might be too emotional or inexperienced in this area). Therefore, we expect that forum usage is positively related to perceived informational and emotional support (H1).

4. Effects of forum activity In addition to the type of social support, engagement level in online communities seems to play an important role in whether individuals receive effective social support [38]. However, existing theories on online support networks do not differentiate between different levels of forum activity or different kinds of users with regard to the intensity of participation. Some users of online forums are active contributors, some post a little, but most members do not post at all [39]. Members of this category are described as lurkers, only following the discussions and browsing the forum without contributing to it [40]. Depending on their posting and participation behavior the more active users can be categorized in various user types [41]. In a systematic review of studies on online health communities Carron-Arthur et al. [42] identified different participation styles based on activity, network and content features. So far, they found no consistent participatory style across different health communities apart from high-engaged users with a high post frequency [42].

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

While online communities can be helpful for individuals who only read but do not actively participate or contribute [23], evidence indicates that active participants may benefit more from online social support than their passive counterparts. Several studies for other health topics [43,44,38] have found a higher degree of intensity of participation in online support activities (e.g., the number of posted messages) to be significantly associated with higher perceived social support, with directly requesting support and writing messages helping members of health forums to deal with stressful situations, improve self-efficacy, enhance feelings of empowerment, and increase overall well-being [45,38]. There are several possible explanations as to why active participants get more out of health forums than passive members. One explanation could be that active users ask for individually relevant information and write about their specific situations, prompting tailored support from other members of the online forum [38] and therefore reach optimal matching of social support [30]. This assumption is supported by a computer-aided content analysis using machine learning models to analyze how specific content in online cancer communities elicits certain types of support. While posts containing self-disclosure about emotional negative events initiated primarily answers with emotional support, asking questions increased the provision of information support [46]. A second reason for enhanced social support through active participation might be that simply the act of writing about personal issues might be beneficial for some people [47]. A third reason could be that giving social support to other forum members helps individuals to increase their own ability to cope with health problems [1]. In summary, the research presented in this section has mainly shown that, in addition to passive reception, active participation maybe a prerequisite for participants of online communities to perceive social support. This leads the second hypothesis, that perceived social support is positively associated with forum activity (H2). Hwang et al. [4] and Ballantine and Stephenson [23] were among the first to describe the role of social support in weight loss networks. Hwang et al. [4] combined a content analysis of the forum and surveys with community members to identify core themes of emotional and informational support. These authors therefore concluded that social support plays a prominent role in participants’ weight loss efforts. Ballantine and Stephenson [23] explored how members of a Weight Watchers Facebook page generated emotional or informational social support by using three different communicating styles. Their study demonstrated that social support can be received by either active or passive members who visit regularly, but not by casual users of the forum. Their results indicated that informational support was higher for passive recipients than for active recipients, while emotional support did not vary between passive recipients and active supporters of the Weight Watchers Facebook page. However, the authors only compared different groups of users and did not directly test associations of activity level with the two different forms of social support (i.e., informational versus emotional support). Despite these promising initial results regarding online support groups for weight loss, it remains unclear whether these effects also apply for online support groups for individuals facing obesity. The focus of these two types of forums is different in at least two ways, although overlaps do exist. First, members of weight loss communities also include many people who are not overweight or obese [4]; these individuals presumably have less need for emotional support, because they might experience their weight as more controllable than do obese individuals [30]. Second, as the central goal in weight loss communities (and perhaps even in commercial communities such as the Weight Watchers Facebook page) is weight loss [23], other crucial aspects of obesity (such as coping with discrimina-

3

tion or with comorbidities) might not be addressed by members [4]. Although it does seem plausible that both informational and emotional social support can be increased through active participation in online forums [38], it is still unknown if and how informational or emotional support are associated with different levels of forum activity for this specific disease. Therefore, we wanted to know if active participation had different effects on informational and emotional support in online communities for obese individuals (RQ1). Different factors may also influence the effect of forum activity on social support. One of these factors seems to be gender. Content analyses of health forums—mainly cancer support groups—have shown differences in how men and women utilize support groups: Overall, while females prefer support groups that focus on emotional support, males appear to have a preference for information-oriented support groups [48]. Another factor might be age. Older adults use the internet less frequently than younger users to obtain health information [49], which would logically influence their forum activities and the support they could receive through those channels. A frequently reported factor is healthrelated characteristics, such as perceived health status [50] or certain disease-specific crisis situations [38] that influence how often and intensively users post messages. Lastly, differences in media literacy or media usage can affect activity levels in healthrelated online forums [43,50], with [43] finding that patients who visited an online community more frequently and spent more time there perceived greater social support from the community members. Additionally, Ballantine and Stephenson [23] found that duration of membership was positively correlated with forum activity in a weight loss support group. Therefore, we wanted to know which demographic, obesityrelated, and media-related factors influence the effects of active participation in online forums for obese persons on perceived informational and emotional support. Research Question 2 (RQ2) asks whether forum activity will have a stable correlation with the outcomes of interest after controlling for age, education, gender, Body Mass Index (BMI), treatment status, and medical co-morbidities, as well as the frequency, duration, and length of forum usage.

5. Methods 5.1. Participants and procedure An online-based questionnaire addressing users of online selfhelp groups for obese persons was distributed from June 12, 2014 to July 13, 2014 via several designated German platforms (e.g., www.adipositas24.de, www.adipositas-portal.de). Of the 324 surveys that were submitted, 72 were discarded due to systematic response patterns, completing the questionnaire in fewer than 190 s (the minimum cut-off), or a failure to fully complete the questionnaire, leaving a final sample of 252 surveys (although additional surveys were later discarded, see below). Participants’ average BMI was 37.97 kg/m2 (SD = 10.04), ranging between 21.8 (a normal weight according to the 2016 WHO classification) and 70.49 (Obese; Class III). About 73.5% of participants could be classified as obese (BMI ≥ 30), and 38.9% as morbidly obese (BMI ≥ 40). A small proportion of participants (6%) were of a normal weight (BMI < 25). We decided to exclude these participants from the sample as we wanted to analyze correlations of participation in online forums for overweight and obese persons only. The mean age of the final sample (N = 230) was 44.1 years (SD = 11.2), ranging between 21 and 71 years, and the overwhelming majority were female (88.7%). All materials (including the questionnaire and the recruitment message) were given in German, pretested with members of the target

4

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

group (N = 7) and validated by experts to ensure understandability, ethical approval, and medical correctness. 5.2. Measures 5.2.1. Forum usage Online community usage was measured by asking the following questions: (1) “How long have you been a member of the obesity online community?,” (2) “How often do you visit the obesity online community?,” and (3) “How much time did you spend with the online community during your last visit?”. These questions were based on research by Oh and Lee [43]. 5.2.2. Forum activity Participants’ forum activity was measured using five items [23] that cover both passive (e.g., “I use this page by observing discussions that are taking place”) and active communication styles (e.g., “If I have advice to give with regard to what someone has posted, I will comment”) on a 5-point Likert scale from 1 = “strongly disagree” to 5 = “strongly agree.” Items for active communication styles include asking a question, expressing an opinion, and giving advice to cover different types of forum activity. All passive items (preference for observation and no interactive observation) were reversed, so that the higher score indicate higher activity (␣ = 0.75). We decided for a unidimensional operationalization that is based on users’ perception of their posting behavior as this is the most frequent form in studies of online health support groups to predict health outcomes [42]. 5.2.3. Perceived social support The items for perceived social support were based on research by Ballantine and Stephenson [23] and Oh and Lee [43], who examined social support in online communities. Perceived informational support was assessed using six items (e.g., “Members will offer me necessary information when I face disease-related problems”) and perceived emotional support with seven items (e.g., “Community members encourage me when I hesitate to make up my mind’). All items were rated on a 5-point Likert scale from 1 = “strongly disagree” to 5 = ‘strongly agree.’ The mean score for informational support was 3.78 (SD = 0.81; ␣ = 0.86) and the mean score for emotional support was 3.68 (SD = 0.93; ␣ = 0.92). 5.2.4. Further characteristics Respondents were asked to provide their gender, age, education level, and BMI, as well as any co-morbidities and their treatment status. They stated, whether they have further diseases (e.g., diabetes or high blood pressure) or therapeutic treatment. Data were analyzed using SPSS (Version 22), and descriptive statistics were used to portray sample characteristics and health literacy levels. Hierarchical regression analyses were used to examine predictors of informational and emotional support, controlling for demographic, obesity-related, and media-related characteristics. 6. Results About three quarters (77.6%) of the participants visited their online self-help group at least once a day, and 98.2% at least once a week; an average visit was about 26 min long (SD = 26.4). About two thirds of the respondents had first visited the forum more than one year ago, and 19% had first visited more than five years ago. These results of forum usage, combined with very positive evaluations of their specific forums underline the relevance of these online communities in the daily lives of respondents. The mean index score for forum activity was 3.51 (SD = 0.84), while the means for individual items of the forum activity scale ranged from M = 2.93 (SD = 1.27) to M = 4.00 (SD = 1.12). Sharing

their knowledge and opinions with other members of the community seemed to be more relevant for participants than asking questions or posting messages (see Table 1). Principal components analysis (with Varimax rotation) was used to assess the underlying structure of the measures used for social support. Based on this analysis, a two-factor solution emerged. These factors explained 65.2% of the variance, with all eigenvalues being over 1, all items loading heavily onto one of the factors, and all factors being easily interpretable. Table 2 lists the factor loadings for each of the scale items: The dimension of informational support was found to contain six items, and emotional support contained seven items. The first hypothesis predicted that mere forum usage would increase perceptions of informational and emotional support. Our results indicate that members who visited the obesity online communities daily perceived higher informational and emotional support than less frequent users. There was a significant difference in the informational support scores between daily users (M = 3.87, SD = 0.81) and less frequent users (M = 3.50, SD = 0.75; t(226) = −2.91, p = 0.004), as well as in the emotional support scores between daily users (M = 3.77, SD = 0.92) and less frequent users (M = 3.38, SD = 0.90; t(226) = −2.66, p = 0.008). The other variables of forum usage—length of membership and duration of last forum usage—were not associated with the social support measures. Therefore, H1 could be only supported for one of the three measures of forum usage. The second hypothesis H2 predicted an association between social support and forum activity. In line with our expectations, participants’ activity on obesity online communities was related to both informational and emotional support. The bivariate correlations between the items of forum activity and the indexes of informational and emotional support were significantly positively correlated for nearly all items. Only the item which indicated observing instead of posting messages was not associated with informational support. Thus, H2 was supported for all but one activity item. Correlations were higher for emotional support for all items, excluding those items that referred to specific, concrete information (see Table 1). To address the two research questions, two hierarchical regression analyses were performed (one for informational and one for emotional support), with seven variables entered in three separate blocks: demographic variables, obesity-related variables, and forum usage measures. Forum activity was then entered as a forth block. Table 3 reveals that after controlling for demographics, obesity-related characteristics, and forum usage, forum activity explained 26% of the variance of perceived informational support and 36% of the variance of perceived emotional support. With regard to Research Question 1, the findings of the regression indicate, that both informational and emotional support are influenced by forum activity. However, this influence seems to be more important for emotional support, as here the change in explained variance (R2 values) for the effect of forum activity was considerably larger than for informational support (see Table 3). Research Question 2 pertained to the role of forum activity in relation to demographic and obesity-related characteristics. Several factors appear to influence social support. Co-morbidities were significantly associated with informational and emotional support, as obese persons without co-morbidities perceived themselves as receiving more social support through their online communities than did those with co-morbidities. Treatment status was also a relevant variable for the perception of informational support, as those currently receiving treatment perceived informational support to be higher than those who were not receiving treatment. Age had a negative correlation with emotional support but not informational support, with older members of obesity online groups feeling less

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

5

Table 1 Means, Standard Deviations, and Correlations for Items of the Forum Activity Scale. Item

M

SD

rInSu

rEmSu

I prefer to observe rather than post messages on this page.a I like to express my opinions on this page. If I have a weight-related question I want answered, I will post a question on this page to get a response. If I have advice to give with regards to what someone has posted, I will comment. I do not interactively communicate on this page as I have nothing to contribute. a

2.93

1.27

0.05

0.22

**

2.98

1.14

0.26

**

0.42

**

3.61

1.33

0.60

**

0.59

**

4.00

1.12

0.29

**

0.45

**

4.02

1.02

0.17

*

0.26

**

Note: a = items were reversed, so that higher scores indicate higher activity. InSu = Informational Support, EmSu = Emotional Support; * p < 0.05, ** p < 0.01. Items from [23]. Table 2 Factor Loadings for Scale Items of Social Support. Item

Emotional Support

This page provides me with effective weight loss information. I use this page to gain information about how I should be exercising to lose weight. If I have a question related to losing weight, I can usually find the answers on this page. I find out valuable dietary information on this page. Members will offer me necessary information when I face disease-related problems. Community members will show me how to cope with a difficult situation that is being faced. If I post my concerns on the board, members will try to help me. I receive caring and helping from the community members. When I feel depressed, members will try to understand me and make me feel better. I gain a feeling of acceptance from using this page. I feel that members of this page care about me as they can relate to what I am experiencing. Community members encourage me when I hesitate to make up my mind. Texts posted on the board give me a feeling of intimacy. Variance explained (percentage)

Informational Support 0.79 0.71 0.78 0.70 0.63 0.63

0.88 0.85 0.83 0.79 0.81 0.67 0.57 54.36

10.80

Note: Items from [23] and [43].

Table 3 Influence of Demographics, Obesity-related Characteristics, Forum Usage, and Forum Activity on Informational and Emotional Support. Informational Support

Emotional Support

Block 1: Demographics Age Gender Education R2 Change

−0.30 1.04 −0.98 0.02

−2.11 0.02 −0.63 0.03

Block 2: Obesity-Related Characteristics BMI Comorbidities Treatment R2 Change

−0.14 −2.83 2.31 0.03

Block 3: Forum Usage Forum membership Frequency of visits Visit duration R2 Change

−1.68 1.47 2.46 0.08

Block 4: Forum Activity Forum activity R2 Change Adjusted R2

6.67 0.17 0.26

** *

* ** *** ***

0.12 −2.82 1.19 0.03

*

**

−0.36 0.62 0.23 0.05 9.52 0.36 0.36

*** ***

Note: N = 194. Hierarchical Regression, gender (1 = male, 2 = female), education (0 = low, 1 = high, i.e.; at least university-entrance diploma), BMI = Body Mass Index, therapeutic treatment (0 = no, 1 = yes), co-morbidities (0 = no, 1 = yes). Figures are standardized beta coefficients in the last step of hierarchical regression. * p < 0.05, ** p < 0.01, *** p < 0.001.

emotionally supported by the community but not less informed. BMI did not contribute significantly to any of the models. 7. Discussion This study was designed to explore the role of active participation in online self-help groups for obese persons. Our results

supported the assertion that the forum usage and forum activity influence perceived emotional and informational support. These results indicate that obesity-specific online self-help groups represent an important way for obese persons to receive disease-specific support, in addition to their offline networks. Active participation in these online communities was found not only to generate a higher perception of social support (especially emotional support),

6

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

but this greater perceived support was independent of mere forum usage. This confirms findings by Barak and Dolev-Cohen [51] and by Tanis et al. [38], who reported that participants who are more active in online support groups gain more relevant information and emotional relief than more passive participants of the same groups. Many studies on online support groups have found that forum usage (in terms of duration and frequency) has a positive influence on patients’ perceived social support. However, few studies have separately tested the influence of forum activity on informational and emotional support. We found that both types of perceived support are significantly related to community members’ forum activity, but that the association between forum activity and informational social support is weaker than the relationship between forum activity and emotional support. We assume that informational support is easier to generate by merely being present on the forum and reading the information being presented; this kind of support can also be supplied by formal channels, such as medical staff or websites, and does not necessarily require interaction. However, a correlation analysis of individual items assessing forum activity indicated that informational support can be increased by asking for the specific information that one needs in a certain situation, which would support the assumption that forum activity increases ones’ chances of receiving tailored information from other members of the online forum [38]. Although content analysis indicates that asking questions in a forum primarily elicits responses containing informational support [46], in our survey study asking questions increased both perceived informational and emotional support. It might be the case that receiving individualized informational support also contributes to the perception of emotional support. In contrast, emotional support seems to require different forms of active participation [38], with the items assessing forum activity that ask about sharing opinions and providing social support to others members proving to be more relevant for emotional support than for informational support. The mere act of writing [47] and giving support [1] might explain the influence of forum activity on emotional support, but not on informational support. This assumption is supported by Kim and Lee’s [52] report that the search for information in online support groups for chronic illness was significantly associated with increased problem-focused coping, while providing support to others was positively correlated to emotional relief. In addition to informational and emotional support, posts containing companionship messages (i.e., the exchange about activities of daily life and personal events) might also play an important role for obese adults. Companionship support has been identified as the most important type of support for user engagement in an online health community about cancer groups [53] and might be studied separately in future research. A feeling of sharing and participating in a group can also be beneficial for obese adults, e.g., to increase their physical activity [54]. For the development of theoretical frameworks, this implies that optimal matching Cutrona & Russell, 1990 is not only required for the type of support (i.e., informational or emotional), but also pertains to forum activity. Forum activity seems to increase the chances of optimal matching by prompting tailored support [38]. In light of our findings, future studies should therefore differentiate between the different forms of social support and different forms of forum activity when analyzing processes of social support in online communities. In order to better understand the associations between interactions in online health support communities and health outcomes, surveys could be combined with content analysis based on machine learning and language analysis techniques [46]. These mixed-method studies could also look into other factors (such as the stage of behavior change and of the illness) that might be relevant [41] and identify different user types in online support groups [53].

Unlike other studies [48], gender was not a relevant factor in perceived social support in our study. Although this finding may be due to the high percentage of females in our sample, it has been welldocumented that females are more intensive and frequent users of health-related online information [55,48]. Future studies should therefore analyze whether the documented relationships are also valid for male users. While females tend to prefer support groups that focus on emotional support, males appear to have a preference for information-oriented support groups, and try to receive emotional support primarily from their partners [48]. Based on the self-selective nature of the data collected, the generalizability of the findings of the present study is limited. Individuals who are not members of online social support groups for obese adults were not represented in the data; it can be assumed that the effects of active forum participation would be even stronger if this control group had been included. It can also be assumed that our study recruited only more active forum users; as per Fox’s [39] findings, only a small percentage of users post comments, questions, or information in such forums, meaning that only a small percentage of users would have responded to such an offer to participate in a research study. While the present study provides an initial understanding of how social support is generated on social online networks, some constructs were not included that could shed further light on this topic. For example, dimensions of trust in information (e.g., [49] or considerations regarding self-disclosure [12] could be included to better understand why some members were active and others were not. This would also allow for a categorization of distinctive user types that stretches beyond participants’ posting behavior [41]. Different user types in the online communities might benefit differently from online support groups [53]. A further limitation was that our cross-sectional study, which took place at one point in time, can only establish correlation between variables; it cannot make statements regarding causation, nor predict long-term effects. Both experimental and longitudinal interventions (e.g., [18] using pre–post designs might therefore be more appropriate in future studies to study the long-term development of forum activity and its sustaining effect on informational and emotional support. It can be assumed that forum activity is associated with certain stages of the disease, and that support seekers turn into support providers over time [1]. These findings offer a significant foundation for health-care professionals to make more use of this alternative to traditional offline self-help groups. Offering online self-help groups as a supplement to therapeutic measures could significantly improve patient outcomes, as well as create savings in the health-care sector. In contrast to Das and Faxvaag [12], who regard online forums as particularly useful for those patients who exclude themselves from traditional interventions or experience barriers to making contact with medical experts, our results indicate that online forums for obese persons are more useful for those who are already in treatment. The lack of social support that some patients experience can also be an important reason to provide online health care service [12], especially as the positive effects of online forums seem to be independent of patients’ BMI. According to our study findings, to maximize the therapeutic impact of online forums, patients should be encouraged to ask questions and to answer posts by others members [12]. The fact that patients with co-morbidities felt less supported could indicate that the analyzed obesity forums do not currently meet their specific needs. It might therefore be useful to know what specific problems obese individuals with co-morbidities face and how online forums can better offer social support in this regard. To provide the long-term effects necessary to effectively treat obesity and to support users who are less active in the forum, online communities could be also complemented by smartphone applications that offer

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

Summary points What was already known on the topic: • Online social support increases social support, • Internet health communities allow individuals to interact with peers who share similar health issues and concerns. What this study added to our knowledge: • Engaging in forum activities (e.g., posting messages) beyond passive observation plays a crucial role in obtaining social support via obesity online forums, • Informational support is strongly related to asking specific questions, while emotional support is more strongly related to sharing opinions and providing support for others, • BMI is not relevant for the perception of social support in online forums.

self-monitoring functionalities, and that regularly send personalized supportive text messages [56,57]. Future research could also be linked to content analyses measuring the quality of information for different obesity-related problems. This is a particularly important area to explore as most online self-help groups lack any expert or professional supervision [29]. For health care institutions, proactive web-based care [58] or moderated online support groups might be a fruitful solution to provide expert knowledge and stimulate further forum activities, thereby offering patients social support. Authors’ contributions K. Wasgien contributed substantially to study design, as well as to data collection, analysis and interpretation. D. Reifegerste contributed substantially to study design, data analysis and interpretation, and manuscript preparation. L. Hagen contributed substantially to study design and manuscript preparation. All authors approved the final version of the submitted manuscript. Competing interests None. Acknowledgments The authors thank the administrators of the online communities for assistance with recruiting participants for the surveys. References [1] T.-C. Lin, J.S.-C. Hsu, H.-L. Cheng, C.-M. Chiu, Exploring the relationship between receiving and offering online social support: a dual social support model, Inf. Manage. 52 (3) (2015) 371–383. [2] K.B. Wright, S. Rains, J. Banas, Weak-Tie support network preference and perceived life stress among participants in health-related, computer-mediated support groups, J. Comput. Med. Commun. 15 (4) (2010) 606–624. [3] Y.K. Bartlett, N.S. Coulson, An investigation into the empowerment effects of using online support groups and how this affects health professional/patient communication, Patient Educ. Couns. 83 (1) (2011) 113–119. [4] K.O. Hwang, A.J. Ottenbacher, A.P. Green, M.R. Cannon-Diehl, O. Richardson, E.V. Bernstam, E.J. Thomas, Social support in an Internet weight loss community, Int. J. Med. Inf. 79 (1) (2010) 5–13. [5] A. Barak, M. Boniel-Nissim, J. Suler, Fostering empowerment in online support groups, Comput. Hum. Behav. 24 (5) (2008) 1867–1883. [6] M. Barrera Jr., R.E. Glasgow, H.G. McKay, S.M. Boles, E.G. Feil, Do internet-based support interventions change perceptions of social support?: An experimental trial of approaches for supporting diabetes self-management, Am. J. Commun. Psychol. 30 (5) (2002) 637–654. [7] J.F. Kinzl, Obesity: stigmatization, discrimination, body image, Wien. Med. Wochenschr. 166 (3–4) (2016) 1–4.

7

[8] M. Ng, T. Fleming, M. Robinson, B. Thomson, N. Graetz, C. Margono, . . . Murray, J.L. Christopher, Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013, Lancet 384 (9945) (2014) 766–781. [9] M. Doull, A.M. O’Connor, V. Welch, P. Tugwell, G.A. Wells, Peer support strategies for improving the health and well-being of individuals with chronic diseases, Cochrane Database Syst. Rev. 3 (2005) CD005352. [10] WHO, Obesity and overweight: Fact sheet No. 311, 2016, Retrieved from http://www.who.int/mediacentre/factsheets/fs385/en/. [11] R.M. Anderson, M.M. Funnell, Patient empowerment: myths and misconceptions, Patient Educ. Couns. 79 (3) (2010) 277–282. [12] A. Das, A. Faxvaag, What influences patient participation in an online forum for weight loss surgery?: a qualitative case study, Interact. J. Med. Res. 3 (1) (2014) e4. [13] G. Mensink, A. Schienkiewitz, M. Haftenberger, T. Lampert, T. Ziese, C. Scheidt-Nave, Übergewicht und Adipositas in Deutschland: [Overweight and obesity in Germany], Bundesgesundheitsblatt − Gesundheitsforschung − Gesundheitsschutz 56 (5–6) (2013) 786–794. [14] C.L. Ogden, M.D. Carroll, K.M. Flegal, Prevalence of obesity in the United States, JAMA 312 (2) (2014) 189–190. [15] B.S. Armour, E.A. Courtney-Long, V.A. Campbell, H.R. Wethington, Disability prevalence among healthy weight, overweight, and obese adults, Obesity (Silver Spring, Md.) 21 (4) (2013) 852–855. [16] J. Wardle, L. Cooke, The impact of obesity on psychological well-being. Best practice & research, Clin. Endocrinol. Metab. 19 (3) (2005) 421–440. [17] T.A. Wadden, V.L. Webb, C.H. Moran, B.A. Bailer, Lifestyle modification for obesity: new developments in diet, physical activity, and behavior therapy, Circulation 125 (9) (2012) 1157–1170. [18] J. Harvey-Berino, S. Pintauro, P. Buzzell, E.C. Gold, Effect of internet support on the long-term maintenance of weight loss, Obes. Res. 12 (2) (2004) 320–329. [19] M. Teufel, K. Stephan, A. Kowalski, S. Käsberger, P. Enck, S. Zipfel, K.E. Giel, Impact of biofeedback on self-efficacy and stress reduction in obesity: a randomized controlled pilot study, Appl. Psychophysiol. Biofeedback 38 (3) (2013) 177–184. [20] B.C. Gold, S. Burke, S. Pintauro, P. Buzzell, J. Harvey-Berino, B.C. Gold, J. Harvey-Berino, et al., Weight loss on the web: a pilot study comparing a structured behavioral intervention to a commercial program, Obesity 15 (1) (2007) 155. [21] A. Bandura, The primacy of self-regulation in health promotion, Appl. Psychol. 54 (2) (2005) 245–254. [22] A. Bandura, Health promotion by social cognitive means, Health Educ. Behav. 31 (2) (2004) 143–164. [23] P.W. Ballantine, R.J. Stephenson, Help me, I’m fat! Social support in online weight loss networks, J. Consum. Behav. 10 (6) (2011) 332–337. [24] M.W. Verheijden, J.C. Bakx, C. van Weel, M.A. Koelen, W.A. van Staveren, Role of social support in lifestyle-focused weight management interventions, Eur. J. Clin. Nutr. 59 (2005) S179–S186. [25] R.R. Wing, R.W. Jeffery, Benefits of recruiting participants with friends and increasing social support for weight loss and maintenance, J. Consult. Clin. Psychol. 67 (1) (1999) 132–138. [26] E.S. Anderson, R.A. Winett, J.R. Wojcik, Self-regulation, self-efficacy, outcome expectations, and social support: social cognitive theory and nutrition behavior, Ann. Behav. Med. 34 (3) (2007) 304–312. [27] K. Elfhag, S. Rossner, Who succeeds in maintaining weight loss? a conceptual review of factors associated with weight loss maintenance and weight regain, Obes. Rev. 6 (1) (2005) 67–85. [28] G. Eysenbach, M. Powell John Englesakis, C. Rizo, A. Stern, Health related virtual communities and electronic support groups: systematic review of the effects of online peer to peer interactions, BMJ: Br. Med. J. 328 (7449) (2004), 1166–1160. [29] M. Tanis, Online social support groups, in: A.N. Joinson, K.Y.A. McKenna, T. Postmes, U.-D. Reips (Eds.), The Oxford Handbook of Internet Psychology, Oxford University Press, Oxford, 2007, pp. 139–153. [30] J.W. Turner, J.A. Grube, J. Meyers, Developing an optimal match within online communities: an exploration of CMC support communities and traditional support, J. Commun. 51 (2) (2001) 231–251. [31] H.S. Wentzer, A. Bygholm, Narratives of empowerment and compliance: studies of communication in online patient support groups, Int. J. Med. Inf. 82 (12) (2013) e386–94. [32] K.C. Eichhorn, Soliciting and providing social support over the internet: an investigation of online eating disorder support groups, J. Comput. Mediated Commun. 14 (1) (2008) 67–78. [33] B.D. Loader, S. Muncer, R. Burrows, N. Pleace, S. Nettleton, Medicine on the line? Computer-mediated social support and advice for people with diabetes, Int. J. Soc. Welfare 11 (1) (2002) 53–65. [34] P.K.H. Mo, N.S. Coulson, Exploring the communication of social support within virtual communities: a content analysis of messages posted to an online HIV/AIDS support group, CyberPsychol. Behav. 11 (3) (2008) 371–374. [35] S.H.M. Roffeei, N. Abdullah, S.K.R. Basar, Seeking social support on facebook for children with autism spectrum disorders (ASDs), Int. J. Med. Inf. 84 (5) (2015) 375–385. [36] D.O. Braithwaite, V.R. Waldron, J. Finn, Communication of social support in computer-mediated groups for people with disabilities, Health Commun. 11 (2) (1999) 123–151.

8

D. Reifegerste et al. / International Journal of Medical Informatics 101 (2017) 1–8

[37] N.S. Coulson, Receiving social support online: an analysis of a computer-mediated support group for individuals living with irritable bowel syndrome, CyberPsychology Behav. 8 (6) (2005) 580–584. [38] M. Tanis, E. Das, M. Fortgens-Sillmann, Finding care for the caregiver? Active participation in online health forums attenuates the negative effect of caregiver strain on wellbeing, Communications 36 (1) (2011) 51–66. [39] S. Fox, The Social Life of Health Information 2011, 2011, Retrieved from http://www.pewinternet.org/2011/05/12/the-social-life-of-healthinformation-2011/. [40] Cornelia F. van Uden-Kraan, C.H.C. Drossaert, E. Taal, E.R. Seydel, A.F.J. van de Laar Mart, Self-reported differences in empowerment between lurkers and posters in online patient support groups, J. Med. Internet Res. 10 (2) (2008). [41] J. Huh, B.C. Kwon, S.-H. Kim, S. Lee, J. Choo, J. Kim, J.S. Yi, et al., Personas in online health communities, J. Biomed. Inform. 63 (2016) 212–225. [42] B. Carron-Arthur, K. Ali, J.A. Cunningham, K.M. Griffiths, From help-seekers to influential users: a systematic review of participation styles in online health communities, J. Med. Internet Res. 17 (12) (2015) e271. [43] H.J. Oh, B. Lee, The effect of computer-mediated social support in online communities on patient empowerment and doctor–patient communication, Health Commun. 27 (1) (2012) 30–41. [44] M. Tanis, Health-related on-line forums: what’s the big attraction? J. Health Commun. 13 (7) (2008) 698–714. [45] H.J. Oh, C. Lauckner, J. Boehmer, R. Fewins-Bliss, K. Li, Facebooking for health: an examination into the solicitation and effects of health-related social support on social networking sites, Comput. Hum. Behav. 29 (5) (2013) 2072–2080. [46] Y.-C. Wang, R.E. Kraut, J.M. Levine, Eliciting and receiving online support: using computer-aided content analysis to examine the dynamics of online social support, J. Med. Internet Res. 17 (4) (2015) e99. [47] J.W. Pennebaker, Writing about emotional experiences as a therapeutic process, Psychol. Sci. 8 (3) (1997) 162–166. [48] P.K.H. Mo, S.H. Malik, N.S. Coulson, Gender differences in computer-mediated communication: a systematic literature review of online health-related support groups, Patient Educ. Couns. 75 (1) (2009) 16–24. [49] L.M.S. Miller, R.A. Bell, Online health information seeking: the influence of age, information trustworthiness, and search challenges, J. Aging Health 24 (3) (2012) 525–541. [50] N. Xiao, R. Sharman, H.R. Rao, S. Upadhyaya, Factors influencing online health information search: an empirical analysis of a national cancer-related survey, Decis. Support Syst. 57 (2014) 417–427.

[51] A. Barak, M. Dolev-Cohen, Does activity level in online support groups for distressed adolescents determine emotional relief, Couns. Psychotherapy Res. 6 (3) (2006) 186–190. [52] J.-N. Kim, S. Lee, Communication and cybercoping: coping with chronic illness through communicative action in online support networks, J. Health Commun. (2014) 1–20. [53] X. Wang, K. Zhao, N. Street, (2014). Social support and user engagement in online health communities. In X. Zheng, D. Zeng, & H. Chen (Eds.), Lecture Notes in Computer Science: Vol. 8549. Smart Health. International Conference, ICSH 2014, Beijing, China, July 10-11, 2014. Proceedings (Vol. 8549, pp. 97–110). Cham: Springer International Publishing. [54] D.N. Cavallo, J.D. Brown, D.F. Tate, R.F. DeVellis, C. Zimmer, A.S. Ammerman, The role of companionship, esteem, and informational support in explaining physical activity among young women in an online social network intervention, J. Behav. Med. 37 (5) (2014) 955–966. [55] S. Fox, M. Duggan, (2013). Health online 2013: Pew Internet & American Life Project Washington, DC. [56] A.G. Arens-Volland, L. Spassova, T. Bohn, Promising approaches of computer-supported dietary assessment and management—Current research status and available applications, Int. J. Med. Inf. 84 (12) (2015) 997–1008. [57] J.Y. Kim, K.H. Lee, S.H. Kim, K.H. Kim, J.H. Kim, J.S. Han, W.K. Bae, et al., Needs analysis and development of a tailored mobile message program linked with electronic health records for weight reduction, Int. J. Med. Inf. 82 (11) (2013) 1123–1132. [58] T. Quast, G. Nöcker, (2015). Social media: forum webcare as a proactive information strategy in health promotion: Results of a pilot project by the BZgA. Bundesgesundheitsblatt, Gesundheitsforschung, Gesundheitsschutz, 966–975. [59] T. Lehnert, A. Konnopka, H.-H. König, Übergewicht und Adipositas als Public Health-Problem [Overweight and Obesity as a public health problem], Public Health Forum 24 (2) (2016) 135–138. [60] C.E. Cutrona, D.W. Russell, Type of social support and specific stress: toward a theory of optimal matching, in: B.R. Sarason, I.G. Sarason, G.R. Pierce (Eds.), Social support: An Interactional View, John Wiley & Sons, Oxford, 1990, pp. 319–366.