back to projects


Online Support Groups for People with 

Diabetes:  Do They Have an Effect on 

Personal and Group Identity?


Sarah A. Novak


The University of Texas at Austin


Health Psychology Final Project




            Research demonstrating the mental and physical benefits of support groups for patients with chronic illnesses has been inconclusive.  This study investigated possible identity benefits, both for the individual and the group.  People with diabetes who participate in online support groups were compared to people with diabetes who do not seek online support.  A modified disease identification questionnaire was used to examine how the participants identified with their illness and labels associated with diabetes.  A social identity profile compared the groups on centrality, internal regard, external regard, and sense of belonging in terms of being diabetic.   It was thought that those who get support from peers online would identify themselves more with diabetes and have higher scores for centrality and sense of belonging than those who do not.  Results from the questionnaires supported these hypotheses, and also showed that participants in support groups identified with a wider range of illness labels than non-participants.

Online Support Groups for People with Diabetes:  Do They Have an Effect on Personal and Group Identity?

            Diabetes has been a common focus of research in health psychology for many reasons.  Both forms of the disease (type I, or insulin dependent, and type II, or non-insulin dependent) can occur with a sudden onset, require immediate changes in lifestyle, and necessitate strict maintenance and self-control.  Diabetes is also associated with serious physical complications such as blindness, renal failure, and loss of feeling in the extremities (Hampson, 1997).  These physiological issues can result in emotional distress for patients as they struggle to accept the radical changes that a diagnosis of diabetes requires, and maintain these changes conceivably for the rest of their lives (Beeney, Bakry, & Dunn, 1996).  Disruption of self-concept, self-esteem, and world-view are only a few of the challenges people with chronic illnesses face (White, Richter, & Fry, 1992).  This investigation will focus on the individuals’ subjective perception of diabetes as a part of their self-concept and how this relationship is affected by participation in support groups.  Participants will be assessed in terms of identification with diabetes, illness centrality, and sense of belonging among other people with diabetes.

            The theory of illness representations asserts that people socially construct common sense models that express their thoughts, beliefs, and feelings about diseases that they may or may not have experienced.  The basic attributes of illness representations are symptoms, labels, causes, timelines, consequences and controllability (Levanthal & Diefenbach, 1991).  These representations have serious implications for people’s mental and behavioral responses to illness.  Sarah E. Hampton (1997) has articulated the significance and pitfalls of illness representations for people with diabetes.  Specifically, symptom experience is not typically related to physiological needs, beliefs about initial cause are not related to daily treatment, the timeline and consequences are understood to be grim and indefinite since there is no cure, and treatment is therefore focused on management rather than the cure.  Diabetes highlights the limitations of illness representations.  It is an extremely difficult disease to conceptualize, and this poses problems for patients trying to understand and cope with their illness.

            Society as a whole has trouble understanding chronic illnesses.  Individuals with diabetes also face the challenge of stigmatization and social rejection.  Physical illness can be viewed as deviation from the norm, and this leads to the conceptualization of illness as deviance.  Research has shown that factors including the severity of the illness and perceived behavioral causality led to greater social distance and rejection (Crandall & Moriarty, 1995).  Depending on one’s concept of diabetes, it can fulfill either or both of these factors.  In the present investigation, illness stigmatization will be assessed with the Janel D. Seagal’s Social Identity Profile (manuscript in preparation).  Two of the subscales included measure internal and external regard; these will reflect the individual’s perceptions of themselves as people with diabetes and others’ attitudes and treatment of people with diabetes.  One of the more insidious effects of society’s fear of severe illness is an increasing alienation of the patient from his or her own body (Scheper-Hughes & Lock, 1991).

Social support has been shown to help mediate the emotional stress, confusion, and overall negative experiences with diabetes.  It entails intimacy, integration, nurturance, reassurance, and informational aid (White et al., 1992).  La Greca, et al. (1995) compared social support provided by family and friends of adolescents with diabetes and found that these support systems operated in different domains.  While family members were helpful in supporting daily maintenance behaviors, friends provided more emotional support.  These researchers have suggested including families and peers in the education process in order to enhance this network of support.  Other studies advocate forming groups to provide consistent education as well as emotional support for patients and their families (McKelvey & Borgersen, 1990).  Thus, social support can also come in the form of support groups of similar peers.

  In recent years, support groups in which people with the same disease discuss their common experiences have become much more prevalent (Jacobs & Goodman, 1989).  Recent work by Davison and Pennebaker (1997) has indicated that people are increasingly joining online support groups through the Internet.  Both types of support groups allow individuals to seek and give advice and support to others who are more familiar with their situation.  Studies have shown that participation in support groups is most helpful when used in conjunction with, and not as a substitute for medical treatment (Gilden, et al., 1992).  However, investigations of the effects of support groups have been inconclusive in determining their potential benefits.  Boey (1999) demonstrated that a general social network was more beneficial than diabetic-specific support in mediating symptoms of depression.  Other research found that a structured support group helped elderly people with diabetes maintain better physiological measures initially, but after 48 weeks, the participants were no different or worse than the control group (Robison, 1993).  Obviously, more research is needed to help elucidate the effects of participation in illness support groups.

The goal of the present study is to examine the relationship between illness, identity, and group membership.  Thaigarajan (1998) showed that acceptance of diabetes as a fact of life had a positive effect on quality of life but did not affect metabolic control.  Other researchers predict health improvements if patients are able to acknowledge and integrate the emotional responses to illness into a larger framework of identity and self-understanding (Davison & Pennebaker, 1996).  These ideas suggest that people who can incorporate their illness into their self-concept will be better off mentally and physically.  In addition, studies by Turnbull, Miller, and McFarland (1990) support the notion that distinctive attributes compared to the population often become central components of self-image.  They also reported that the perception of a shared identity creates bonds between people, and this can result in a greater sense of community within the group.  Whether individuals with diabetes agree with these theories remains to be seen.  In the present study, measures of disease identification, centrality, and sense of belonging will be used to examine these results in terms of the experiences of people with diabetes.

The hypotheses tested were that people with diabetes who participate in online support groups would identify more with their illness according to the disease identification questionnaire.  This is anticipated because not only do support group members seek out similar peers to interact with, they also spend more time communicating with people who can relate to and sympathize more completely than other confidants.  The diabetes-related labels that participants identify with will provide another type of identification, and support group members are expected to be more positive overall than non-members because of the support that they receive from peers.  Support group participants are also expected to have a higher degree of illness centrality and sense of belonging to the diabetic community as determined by the social identity profile because of the distinctiveness of their illness relative to the population as a whole.





            The participants were 12 individuals with diabetes.  Online support group participants (4 women, 2 men, 5 type I, 1 type II, mean age = 38.16) were recruited from Internet newsgroups dealing with diabetic support.  Participants not involved in support groups (1 woman, 5 men, 4 type I, 2 type II, mean age = 33.5) were recruited via email messages from acquaintances.  All participation was strictly voluntary.


            Two unpublished surveys were modified for use in this experiment.  The Disease Identification Questionnaire was originally created by Pennebaker and the Social Identity Profile was created by Seagal (1999).  A sample of the modified and combined questionnaires is provided in the Appendix.


            Messages were posted on several different Internet newsgroups in order to recruit online support group participants.  All of the selected newsgroups focused on the topic of diabetes support.  Electronic mail messages were sent to various listserves within the university community to recruit people with diabetes who were not involved in support groups.  Volunteers who responded to either form of message were emailed copies of the questionnaire and told where to email or mail them back to the experimenter. 



            Multiple analyses of variance were used to compare the two groups of participants on each measure.  An alpha of .05 was used in each test.

            The disease identification questionnaire had several significant results as shown in Figure 1.   In response to question one, online group members reported that they think of themselves as people with diabetes significantly more than non-group members, F(1, 10) = 5, p = .049.  On question four, group members said that diabetes is significantly more of a problem in their lives, F(1, 10) = 10.321, p = .009.  For question seven, online support group participants reported that diabetes is a core part of who they are significantly more than the participants who are not support group members, F(1,10) = 10.74, p = .0084.  All other questions in this section were not answered significantly differently by the two groups of participants.

            The disease identification questionnaire also asked participants how much they identified with particular labels related to diabetes (see Figure 2).  The label that both groups identified with the most was “a person living with diabetes.”  The online support group participants had less extreme choices than the non-group participants.  The support group members also identified significantly more with both “a diabetes victim,” F(1, 10) = 5.68, p = .038, and “a diabetes survivor,” F(1, 10) = 6.20, p = .032.

            The social identity profile consists of four subscales on which the groups were compared as shown in Figure 3.

            The groups were not significantly different in internal and external regard.  The online support participants were nearly significantly different in centrality, F(1,10) = 4.86, p = .052.  The anova for sense of belonging revealed a much greater degree of significance, with online support participants scoring much higher than the non-support group, F(1, 10) = 19.45, p = .001.


            The results of this investigation support some hypotheses while disconfirming others.  Analysis of the disease identification questionnaire revealed several significant results.  It was expected that participants in support groups would identify with their illness more than those who do not participate in groups, and responses to two of the questions support this hypothesis.  Support group members reported that they think of themselves as people with diabetes more and consider diabetes to be a core part of who they are more than non-group members.  These results seem intuitive – if someone is interacting with other people with diabetes more, particularly when the focus of communication is experiences with the illness it is logical that the individual would consider that illness more a part of him or herself.  The third significant result, in which support group participants said that diabetes is more of a problem in their lives seems less obvious.  One possibility is that people who are having a harder time with their illness are more motivated to seek information and support from similar others.  It is also possible that because support group participants are confronted with both their own problems and the problems of fellow members they are more troubled by the disease (Pennebaker, 1997).  It could also be true that even though support group members are expressing their difficulties to others, they may not be doing so in a productive way.  Pennebaker (1997) suggest that ruminating over problems is not an effective coping strategy, which could explain why these individuals view their illness as more of a hardship than others.

            Other results to the disease identification questionnaire are notable for their insignificance.  No significant differences were reported on factors such as how much the participants talk to others about their experience, how important it is for friends to either know or not know about their illness, how important it is to talk to others with similar experience, and even how willing they would be to participate in a support group.  Perhaps further testing with a larger subject pool and knowledge of how active each participant is within the groups would produce more differences.  Otherwise, these results seem to indicate that there may not be that much difference in the identity of those who participate in support groups and those who do not.  One result that should not go unnoticed is that there is a nearly significant difference between how much each group expects participating in a support group for people with diabetes would impact their lives.  This may indicate a self-fulfilling prophecy; those who expect to benefit from support groups might interpret themselves as doing so, whether or not there is a real effect.  However, because no physiological or behavioral indicators of health are being considered in this investigation, it is not possible to know whether there is a difference between the participants’ perceptions and the real effects.

            When asked what labels they identify with, both groups rated “a person living with diabetes” the highest.  This choice was included to represent the gray area between the extremes of classification as “a victim” or “a survivor.”  Though it was expected that support group members would identify more with the term “survivor,” it was surprising that they also identified more with the term “victim.”  As before, this could reflect these people’s motivations for seeking (victims) and giving (survivors) support in the first place.  The majority of participants in both groups identified highly with more than one label, demonstrating that none of these is mutually exclusive.  The support group members showed more even distributions across all of the labels, while the non-members used more extreme scores, mostly using ones and sevens to represent their choices.  Perhaps by writing about their experiences and reading those of others, the support group members are more cognizant of all of their various moods and experiences, while non-support group members instead recall only major episodes.      

            The social identity profile was used to assess possible differences with four subscales.  Internal regard was not significant because it was similarly low for both groups.  This indicates that there are few positive aspects of having a chronic illness, which is hardly surprising.  External regard was not significant because it was similarly high for both groups.  If these individual’s experiences can generalize to others with diabetes, it seems that diabetes does not create perceived stigmatization and social rejection as some researchers have predicted.  However, further tests with measures that focus specifically on illness stigmatization are necessary before drawing this conclusion.   The groups did vary significantly on illness centrality.  For instance, support group participants agreed more with statements on whether being a person with diabetes is central to their sense of who they are, and whether it has to do with how they feel about themselves.  Obviously, illness centrality is basically synonymous with disease identification and previous results are reaffirmed by this measure.  The greatest significance was found in the measurement of sense of belonging.  Support group members agreed much more with statements of whether they could contribute to a community of people with diabetes and whether they feel good around other people with diabetes.  It seems clear that people who feel they can contribute and be a valuable member of a community of people with diabetes would seek out such a community, whether in face-to-face meetings or through online support groups.  What remains to be seen is whether people who do not already have those feelings and desire for community with similar peers would develop them by participating in support groups.

            Implications of this investigation include increased evidence of the value of support groups in terms of helping individuals integrate diabetes into their self-concept and identity, as well as promoting belonging among people with diabetes.  Further experiments could include participants from face-to-face support groups and more participants overall to see whether these results are stable.  If possible, some sort of randomization should be used to protect against self-selection and personal expectations creating the effects.  It would also be interesting to examine writing samples from both support group members and non-members to get a closer look at the type of expression each group creates in regard to illness.  In addition, some form of rating one’s health condition (preferably objectively) could also provide meaningful insight on how identifying with a disease and feeling like a member of a community impact health outcomes.


            Beeney, L.J., Bakry, A.A., & Dunn, S.M. (1996).  Patient psychological and information needs when the diagnosis is diabetes.  Patient Education and Counseling, 29, 109-116.


            Boey, K.W. (1999).  Adaptation to type II diabetes mellitus: Depression and related factors.  International Medical Journal, 6(2), 125-132.


            Crandall, C.S., & Moriarty, D. (1995).  Physical illness stigma and social rejection.  British Journal of Social Psychology, 34(1), 67-83.


            Davison, K.P., & Pennebaker, J.W. (1996).  Emotions, thoughts and healing: After Dafter.  Advances, 12(2), 19-23.


            Davison, K.P., & Pennebaker, J.W. (1997). In K. Petrie, & J. Weinman (Eds.) , Perceptions of Health and Illness (pp. 463-486).  United Kingdom: Harwood Academic Publishers.


            Hampson, S.E. (1997).  Illness Representation and the Self-Management of Diabetes.  In K. Petrie, & J. Weinman (Eds.) , Perceptions of Health and Illness (pp. 323-347).  United Kingdom: Harwood Academic Publishers.


            Gilden, J.L., Hendryx, M.S., Clar, S., & Casia, C. (1992).  Diabetes support groups improve health care of older diabetic patients.  Journal of the American Geriatrics Society, 40(2), 147-150.


            La Greca, A.M., Auslander, W.F., Greco, P., Spetter, D., Fisher, E.B., & Santiago, J.V. (1995).  I get by with a little help from my family and friends: Adolescents’ support for diabetes care.  Journal of Pediatric Psychology, 20(4), 449-476.


            McKelvey, J., & Borgersen, M. (1990).  Family development and the use of diabetes groups: Experience with a model approach.  Patient Education and Counseling, 16(1), 61-67.


            Robison, F.F. (1993).  A training and support group for elderly diabetics: Description and evaluation.  Journal for Specialists in Group Work, 18(3), 127-136.


            Scheper-Hughes, N., & Lock, M. (1991).  The Message in a bottle: Illness and the micropolitics of resistance.  Journal of Psychohistory, 18(4), 409-432.


            Seagal, J.D. (1999).  The Social Identity Profile: How members of socially stigmatized groups think and feel about their group membership.  Unpublished manuscript, The University of Texas at Austin.


            Thiagarajan, K.D. (1998).  Stress, social support, problem-solving coping, acceptance of diabetes and self-management as predictors of metabolic control and quality of life among adults with insulin-dependent diabetes mellitus.  Dissertation Abstracts International: Section B: the sciences and engineering, 59(6-b), 2688.


            Turnbull, W., Miller, D.T., & McFarland, C. (1990).  Population-distinctiveness, identity, and bonding.  In J.M. Olson, & M.P. Zanna (Eds.), Self-inference processes: The Ontario symposium (pp.115-133).  Hillsdale, NJ, USA: Lawrence Erlbaum Associates, Inc.


            White, N.E., Richter, J.M., & Fry, C. (1992).  Coping, social support, and adaptation to chronic illness.  Western Journal of Nursing Research, 14(2), 211-224.


            Wysocki, T., Harris, M.A., Greco, P., Harvey, L.M., McDonell, K., Danda, C.L., Bubb, J., & White, N.H. (1997).  Social validity of support group and behavior therapy interventions for families of adolescents with insulin-dependent diabetes mellitus.  Journal of Pediatric Psychology, 22(5),  635-649.