“Everyone Deserves to Look Normal”:
The Communication of Uncertainty, Support, and Information
in an Online Community for Families Affected by a Vascular Birthmark
Liesbeth Wiering
M.A. Candidate
San Diego State University

Every year, ten percent of all babies in the US are born with a vascular birthmark (Shannon & Marshall, 1997). This means that approximately four hundred thousand newborns show spots and stains on their skin that are often endearingly called stork bites, strawberries, and angel kisses. In ninety percent of the cases, these skin flaws and discolorations disappear within a year. For the other ten percent, things may literally not look so good. The degree to which a person is “marked for life” (Davidow, 2003) will depend on several factors such as the location of the birthmark, how easily it can be obscured, and on the availability and affordability of adequate treatment. Severe physical complications are possible, and people with birthmarks may suffer debilitating social and psychological consequences in a society that doesn’t always treat people who “don’t look normal” respectfully (Shannon & Marshall; Troilius, Wrangsjö, & Ljunggren, 1998, 2000).

The diagnosis, prognosis, and treatment of a birthmark are surrounded by uncertainty. Whether a stain found on a baby’s body is indicative of a birthmark requiring medical attention is not always immediately clear. The complexity and relative rarity of occurrence make it hard for doctors involved in the early stages of diagnosis to make decisions. Moreover, getting proper information is complicated by the fact that no single medical discipline is specialized in vascular birthmarks. People often need to go from one specialist to another (e.g., seeing a dermatologist for one part of the problem, and a plastic surgeon for another), which may add to the confusion (Shannon & Marshall, 1997).

The experience of uncertainty is considered to be a central element in dealing with health and illness (Mishel, 1988, 1990). Related to this idea is the common belief that communication and information acquisition are essential to the management of uncertainty (Babrow, Kasch, & Ford, 1998). People confronted with an acute or chronic disease or disorder may respond to their illness-related uncertainty by looking for information beyond the doctor (Dutta-Bergman, 2005). This may be particularly the case for people affected by a birthmark, since, as noted above, the information in the medical field is scattered over many disciplines.

An important source of illness information can be found in contact with similar others (Brashers, Goldsmith, & Hsieh, 2002). Communication with peers, often taking on the form of support groups, may offer a form of relief and understanding that family, friends, and medical workers are unable to provide. For people with a rare disease it may be difficult to find others with the same illness in their geographical area (Lasker, Sogolow, & Sharim, 2005; Patsos, 2001). One strategy to counteract the isolation of dealing with a rare disease is to establish contact with e-mail discussion groups (Calhoun, 1998). Electronic support groups offer people not only support and knowledge, but also membership to a large, understanding, and experienced community (Galegher, Sproull, & Kiesler, 1998).

While several studies have looked at the communication of information and support in health-oriented online communities and on message boards (e.g., Galegher et al., 1998; Lasker et al., 2005; Macias, Lewis, & Smith, 2005; Muthaseb & Wright, 2005), no study has specifically focused on the role of illness-related uncertainty in online support groups and its communication and management. The present study aims to fill this gap by taking a closer look at how uncertainty is communicated and managed by requests for and receipt of support and information in an online community for families affected by a vascular birthmark. An examination of relevant literature on illness-related uncertainty is followed by an overview of the most common types of vascular birthmarks and their related problems. After reviewing the role of support and information in online communities, several questions are formulated that apply findings from previous research to the case of families affected by a birthmark. This is followed by a description of the methods used for this study and a discussion of the findings.

Confronting Illness and Uncertainty

The fact that people diagnosed with an illness will face significant illness-related uncertainty has been demonstrated in numerous studies (e.g., Babrow, Hines, & Kasch, 2000; Cohen, 1993; Mishel, 1988, 1990). Illness here takes on the larger meaning of “compromised health” and thus also includes conditions such as having a specific disorder, disability or condition that may not necessarily be accompanied by a state of being ill. Mishel (1988) defined uncertainty in illness as the “inability to determine the meaning of illness-related events [that] occur in situations where the decision maker is unable to assign definite values to objects and events and/or is unable to accurately predict outcomes because sufficient cues are lacking” (p. 256).

Illness-related uncertainty may be a result of various phenomena. People may be unable to understand or explain the diagnosis, they may worry about consequences for the future, feel insecure about their bodies, experience distrust of their doctors, or just not know what to do next (Babrow & Kline, 2000; Brashers et al., 2002). Following Mishel’s claims, Babrow et al. (1998; 2000) have stressed the many causes and meanings illness-related uncertainty can have within the context of Problematic Integration (PI) theory. Uncertainty may arise from the complexity of the illness. People may find the illness they are confronted with hard to understand and may be unable to tell what caused it. They may feel uncertain about the quality of the information they have, feel they don’t have enough information, or on the contrary, too much. The outcome of the illness may not be clear, and similarly, people may not know which possible course their illness will take, or what having this illness may mean to their life. However, the evaluation of uncertainty, the extent to which it is experienced as a negative (or a positive) thing, and people’s response to it may predominantly depend on people’s personal desire to know what there is to know (Babrow et al. 2000; Brashers et al.).

In western societies it is generally assumed that people encountering illness, and thus uncertainty, will have a desire to gather information on their condition in order to obtain explanations and make predictions about their situation. They may want to better understand the diagnosis, be able to make well-founded decisions on treatments, or to make adequate predictions about their prognosis (Brashers et al., 2002). Also, patients may want to prepare themselves for what is coming (Hines, 2001).

Closely related to general information gathering and also well-documented is the role of social support in the management of illness-related uncertainty. Brashers, Neidig and Goldsmith (2004) have stressed how supportive others can help with information seeking, provide instrumental support (e.g., assistance with transportation and household tasks), help with the development of coping skills, and give validation and acceptance. According to Ford, Babrow, & Stohl (1996), social support messages can facilitate coping by providing information that contributes to reaching a desired level of uncertainty, by either reducing, maintaining, or increasing it. In some cases, uncertainty is not necessarily experienced as a negative thing. Some patients may look for means to increase uncertainty as it can provide a source of hope (Babrow et al., 1998).

People affected by a vascular birthmark are expected to experience uncertainty, not only because of the complexity and the unpredictability of the development of the condition, but also because their condition is rare, yet often very visible and resulting in severe psycho-social consequences.

Confronting a Vascular Birthmark Diagnosis

The cause of vascular birthmarks is yet unknown. While doctors and medical researchers have assured that vascular birthmarks are not inherited or caused by specific eating habits and activities during pregnancy, many people still believe myths such as “food cravings are linked to birthmarks” (Burgoon & Hall, 1994, p. 103). It may be hard to determine what type of birthmark a baby has. Misdiagnoses, especially in the early stages, are not uncommon. Medical schools do not teach about vascular birthmarks as a specific topic (L. R. Shannon, personal communication, March 19, 2006) and doctors involved in the early stages of diagnosis tend to be relatively ignorant about the potential for complications, resulting in diverse, often inadequately formed opinions with regards to the necessity of treatment (Shannon & Marshall, 1997).

The most common types of vascular birthmarks are hemangiomas, port wine stains (PWS), and arteriovenous malformations (AVM) (National Organization of Vascular Anomalies, 2004; Syed, 1999). Hemangiomas happen in 4 to 10% of all children in the United States, but not all of them require active medical attention. PWS occurs in .33% of births, and the estimated incidence of AVM is .14% (Smith, 2006).

Hemangiomas are blood-filled, benign tumors (Shannon & Marshall, 1997). Present more often among girls and babies with a low birth weight, almost a third of hemangiomas are visible right away, while the others appear within a month after birth. The vast majority occur in the head and neck area, mostly between the eyes and on the lower part of the face, although they can develop anywhere on the body, both internally and externally (Vascular Birthmark Foundation, n.d.a). Hemangiomas grow progressively during the first 12 months of life, after which a process of shrinkage (“involution”) follows that can take several years. Hemangiomas can become quite large, have serious complications such as ulceration and airway obstruction, and leave disfiguring scars. Treatment may be recommended but not always thought to be the best choice. Among the treatment options are: laser therapy, steroid treatment, interferon treatment, and surgical incision (“debulking”) (Shannon & Marshall).

A PWS is a venous malformation that is present at birth and always permanent. PWSs are mostly flat stains that can range in color from light pink to dark purple (Vascular Birthmark Foundation, n.d.a). Most people have heard of PWS or may remember the one on the head of former President of the Soviet Union Mikhail Gorbachev. The majority of PWSs appear on the face. With maturity they may darken, thicken and develop lumps and bumps that bleed easily. PWS can be indicative of larger serious disorders such as Sturge-Weber syndrome, which involves vascular malformations in the brain (Shannon & Marshall, 1997). Since the late eighties, the most common treatment for PWS is laser therapy, aimed at lightening the stains and preventing the further development of PWS-related complications.

AVMs are tangles of abnormally formed blood or lymphatic vessels. They can happen anywhere in the body. Although present at birth, they may not be detected until much later. AVMs in the brain are thought to be the leading cause of stroke for people under the age of thirty (National Organization of Vascular Anomalies, 2004). Visible AVMs in infants are often mistaken for hemangiomas. However, they are importantly different as AVMs do not involute and/or spontaneously disappear. Like PWSs, they are a life-time condition. AVMs grow over time, particularly during times of illness and hormonal changes (e.g., in puberty). The high chance of rupture poses an ongoing risk to health (Vascular Birthmark Foundation, n.d.a ). Among the treatment options are conventional surgery, endovascular therapy, laser surgery, and radio surgery.

The prediction of the course of development of individual hemangiomas, PWSs, or AVMs is surrounded by uncertainty, as no two vascular birthmarks are the same (Shannon & Marshall, 1997). Most often it is impossible to foretell what will happen, which makes the choice of treatment very difficult. Hemangiomas have only a small window of opportunity that allows the intervention of possible disfiguring growth through the administration of steroids. Some hemangiomas may be better off left alone and steroids can produce undesirable side effects. Parents may feel reluctant to put their newborn on such heavy medication. While most PWSs respond well to laser treatment, others do not respond at all. AVM surgery can be dangerous and result in excessive bleeding. Laser treatment can be painful and in children often requires general anesthesia (Shannon & Marshall). The severity of disfigurement people face and the overall consequences that having a vascular birthmark entails can generally only be discovered with the passing of time.

The chance of getting adequate treatment may further be complicated by insurance companies that regard laser therapy in birthmarks as a “cosmetic procedure” and deny patients coverage (McClean & Hanke, 1997). This point of view not only tends to ignore the potential for medical complications but also disregards the possible psychological damage that having a vascular birthmark may cause, which can be equally jeopardizing to general health and may require treatment (and subsequent health insurance coverage!) later on. On this account, Troilius, Wrangsjö, and Ljunggren (1998) claim that early treatment of vascular birthmarks in children can prevent or diminish the severe psychosocial distress that people affected by a birthmark often experience. Based on a survey among 259 people with PWS, Troilius, Wrangsjö, and Ljunggren (2000) revealed that 80% had not accepted their condition and thought their lives would improve if their PWS could be removed without scarring. More than half of the respondents believed others treated them differently because of their PWS and a third was convinced their PWS had prevented them from making friends. This study also found that respondents’ self-esteem improved significantly after even only relatively successful treatment. Similarly, a study aimed to assess quality of life and satisfaction with laser treatment among 25 people with PWS found that those who had more successful treatment consequently showed lower scores on both a helplessness scale and an anxiety/avoidance scale (Schiffner, Brunnberg, Hohenleutner, Stolz, & Landthaler, 2002).

People diagnosed with a vascular birthmark may feel lonely and isolated, and have the idea that they are the only person in the world who looks like they do and are forced to deal with such a condition (see Davidow, 2003). Because of the relative rarity of occurrence, and the tendency of those affected to hide their condition from the public, establishing contact with similar others has often been hard in the past. The arrival of the internet has brought support and relief from isolation for many people struggling with similar conditions.

Finding Support and Information in Online Discussion Groups

Numerous studies on the experience of illness have pointed out the importance of support from people who have been in a similar situation, or from those who are going or have gone through the same experience (e.g., Ainbinder, et al., 1998; Brashers et al., 2002; Davison, Pennebaker, & Dickerson, 2000; Klemm & Hardie, 2002). Increasingly popular among people looking for peers are internet-based support groups. These have also been identified as one of the primary sources for online health information seeking today (Cotten & Gupta, 2004). Research efforts that have looked at attraction to internet support groups and communication in online communities have found that the general advantages that users mention most often are the 24-hour accessibility, immediacy of information, at-home availability, anonymity, privacy, and option of selective participation (see e.g., Cotten & Gupta; Han & Belcher, 2001; Galegher et al., 1998; Klemm & Hardie; Reeves, 2000).

Online support groups provide the exceptional ability to reach and connect those people, for whom the traditional support group structure of face-to-face meetings may not be an option, and for whom alternative ways of information gathering are sparse. This seems to be especially the case for people facing isolation either because they are suffering from a disease that (a) is relatively rare (Lasker et al., 2005; Muthaseb & Wright, 2005), (b) is complex (Patsos, 2001), (c) is misunderstood (Davison, Pennebaker, & Dickerson, 2000; Lasker et al.), (d) is embarrassing, socially stigmatizing or disfiguring (Davison, et al.; Galegher et al., 1998; Muthaseb & Wright; Reeves, 2000), or (e) leaves them homebound (Klemm & Hardie; Reeves).

One advantage of communicating in online support groups is the opportunity of information gathering. For instance, in a content analysis of messages posted to a mailinglist for people with the rare disease primary biliary cirrhosis, Lasker et al. (2005) found an emphasis on sharing experiences and exchanging information with people with similar conditions. According to these researchers, the need for information among people with a rare disease is further motivated by the fact that not only peers, but also competent medical specialists are hard to find. Similarly, Patsos (2001) has stressed that it is common for people with a complex, rare disease “to see many specialists, who often have diverse opinions in diagnosis and treatment. This makes it especially critical for patients to have trustworthy and reliable internet sites to help them clarify uncertainties and misinformation” (p. 805). In an extensive study that compared face-to-face support group use to use of internet support groups, Davison et al. (2000) found that online communities seem to be more oriented around conditions that are poorly understood or generally overlooked by medical research.

A second advantage participants in internet support groups often mentioned is the general lack of feeling judged by others (e.g., Cotten & Gupta, 2004; Muthaseb & Wright, 2005; Reeves, 2000). On this account, in a study that examined HIV-positive individuals’ online coping strategies, Reeves (2005) found that participants used online communication for ‘heavy’ issues that they found difficult to discuss face-to-face, such as death and intimate bodily functions. Equally, Cotten and Gupta (2004) mentioned that the anonymous nature of the internet “allows users to ask awkward, sensitive, or detailed questions without the risk of facing judgment, scrutiny, or stigma, and to do so at their convenience” (p. 1797). In a study of perceived online support group advantages, Muthaseb and Wright (2005) noted that participants may profit from the lack of distracting non-verbal cues “such as appearance” (p. 14). Davison et al. (2000) stressed that for some people virtual support may be especially appealing because “potential social factors like physical attractiveness, vocal characteristics, ethnicity, and social skills are neutralized” (p. 211).

Receiving validation for experiences and symptoms is a third benefit that internet users often express (Dickerson et al., 2006; Lasker et al., 2005; Patsos, 2001; Reeves, 2000). Similar others may recognize symptoms, and make confused people feel better by letting them know that they have gone through the same experience, since “doctors don’t think to tell you all the little things” (Dickerson et al., p. E14). Hearing that others have lived through similar situations can give people the feeling that they are “normal” after all (Reeves).

The availability of online support introduces an extra dimension of social connection and access to information (Davison et al., 2000). Access to multiple perspectives, empowerment, and patient advocacy are other positive aspects of online support groups that participants have mentioned repeatedly (e.g., Muthaseb & Wright 2005; Patsos; Reeves).

All these internet advantages are factors that may help people confronted with a rare or stigmatizing condition, such as a vascular birthmark, cope, surmount isolation, feel less confused, and become competent, knowledgeable communication partners and decision makers in their own illness trajectory. As an example of this mechanism, this research aims to reveal how families affected by vascular birthmarks make use of an online discussion forum, and which types of birthmark-related uncertainty they most often express.

Research Questions

People affected by a birthmark have to deal with a relatively rare, complex condition that is often disfiguring and socially stigmatizing, and that is surrounded by embarrassment, misinformation, and multiple forms of additional uncertainty. Therefore, this group may particularly benefit from participation in an online support group. The present study aims to reveal how families affected by vascular birthmarks make use of the online discussion forum of the Vascular Birthmark Foundation (VBF). Its goal is to identify and classify types of uncertainty that members of an online community for families affected by a birthmark struggle with by (a) content analyzing messages that they posted in a discussion forum, and (b) examining responses to an online survey that respondents linked to through this discussion forum. Particularly, this study intends to examine what types of “illness-related uncertainty” are communicated, in an effort to identify which type of uncertainty is most often reported, by whom, and how other posters respond to these expressions. Therefore, the following research questions were proposed:

RQ1: Which types of uncertainty do people affected by a birthmark most dominantly express on the VBF discussion forum?

RQ2: Who are the people who express uncertainty on the VBF discussion forum? Are they generally new to the group or are they return visitors? Are they generally parents or people with a vascular birthmark themselves? Are they generally female or male?

RQ3: Which types of responses do posters on the VBF discussion forum generally get? How are legitimacy and authority established in these responses?

RQ4: What characterizes the people who responded to the online survey?

RQ5: Which types of uncertainty do people affected by a birthmark most dominantly report in response to an online survey?

RQ6: What benefits do the survey respondents report to get from the communication of the VBF discussion forum?

Related questions that this study aims to answer are: Does uncertainty obtrusively or unobtrusively shine through in the conversations on the VBF discussion forum? Is uncertainty explicitly or inexplicitly mentioned in the messages? When does uncertainty occur most importantly, at what phase of confrontation with a birthmark and/or stage of diagnosis and treatment? How important is the role of health insurance in seeking treatment for vascular birthmarks? Answers to these questions will not only provide a better understanding of a largely ignored area of both medical and sociological research, it may also give the community under study a sense of validation and a tangible role in the quest for a cure.

Method

This investigation focuses on online communication that is exchanged on the web-based discussion forum of the Vascular Birthmark Foundation (VBF). The VBF is “an international charitable organization that provides support and informational resources for individuals affected by hemangiomas, port wine stains, and other vascular birthmarks and tumors, sponsors relevant research and promotes physician education” (Vascular Birthmark Foundation, n.d.b). Its primary objective is to “enable every child born with a vascular birthmark to receive an accurate diagnosis and appropriate treatment” (Vascular Birthmark Foundation, n.d.c). The VBF’s online discussion forum has six condition-specific categories, ranging from “Hemangioma Help” to “Youth and Teen Support.” Over the last 7 years, 2454 members have posted close to ten thousand messages on more than 1800 topics (also known as threads, such as “Pain killers before laser?;” “Can anyone tell me about lip hypertrophy?;” “Knowing that you’re doing the right thing”). Members are roughly divided in two groups: people who have a vascular birthmark themselves, and parents of children with vascular birthmarks.

Instrumentation of the study is two-way: (1) a quantitative content analysis was done of messages posted to the VBF discussion forum, and (2) a mixed-methods online survey was announced in a newly started thread, with a plea for anonymous participation. The following section first discusses the steps that were necessary to assure ethical conduct of the study. Secondly, the sampling and coding methods for the content analysis are explained, followed by an interpretation and justification of the chosen survey method and its included measures.

Ethical Conduct

Permission for the study was requested from the Institutional Review Board (IRB) of San Diego State University, as well as from the founder of the VBF, Linda Rozell Shannon. While the founder of the board welcomed the attention to the vascular birthmark community and its key issues (L. R. Shannon, personal communication, March 19, 2006) and readily approved the research initiative, getting IRB approval proved to be much more difficult. The IRB approved the survey, but, initially, not the use of the online messages for the study. This was surprising and unexpected as several other studies have used similar samples and methods (e.g., Galegher et al., 1998; Lasker et al., 2005), apparently without encountering these problems. Moreover, a case can be made for the opinion that online messages, such as found on the VBF discussion board, are part of the “public domain,” as they are freely accessible to anyone with an internet connection without the need for organization membership. However, the IRB insisted that people who post to such forums do not reasonably expect that their messages will be used for research purposes, and that they should therefore be asked for permission first. Verifying online identity is a complex matter, and tracing back and getting in contact with all people who posted in the sample period seemed a virtually impossible task. The researcher did not consider this as a feasible option.

In a similar regard, Finn (1999) has made mention of the controversy and the lack of consensus that exists in the research community at large on the ethics of using online messages for research purposes. One of the solutions he proposed and applied to his study on help processes in an online self-help group (Finn, 1999) has also been applied to the present study: the abstinence of reporting any identifying information from the messages. This is especially important for this study since, unlike Finn’s study, the name of the message board it has taken its sample from has been disclosed. The researcher decided to do this not only because the organization is of such great importance to the vascular birthmark community, but also because she is of the opinion that for proper research conduct, such a level of transparency is important.

The SDSU IRB eventually permitted the use of the VBF messages for content analysis after ensuring that no identifying information would be present in any of the data files. Also, as mentioned above, I posted a personal message with a link to the online questionnaire. This message included a general notification to users of the VBF discussion forum explaining the researcher’s intentions for the content analysis with a plea for reactions and potential ethical objections (see Appendix C). The few reactions received were all of a positive nature and no posters raised objections to their messages being included in the sample for the content analysis.

Content Analysis Sample and Coding System

The sample for the content analysis part of this study consisted of the messages from those VBF discussion forum threads that had activity during a three-month time period (January1, 2006 through March 30, 2006) in the three major categories on the VBF community discussion forum, “Hemangioma Help,” “Port Wine Stain Help,” and “AVM Help.” This yielded a sample of 143 threads that included a total of 960 messages. The sample consisted of 104 Hemangioma Help threads, 29 Port Wine Stain threads, and 10 AVM Help threads.

The unit of analysis for the first part of the content analysis was the initial message of the thread (N = 143). A database was created in SPSS 11.5. To build a coding system, the messages were read several times and scanned for recurring uncertainty patterns. A total of 28 types of uncertainty were derived from the messages. These were classified under four major uncertainty categories of complexity, information quality, probability, and integration, based on Babrow et al.’s (1998, 2000) meanings of uncertainty (see Table 1). A fifth “non-uncertainty” category was added for those messages that did not contain expressions of uncertainty, but were used to communicate with VBF people for other reasons (e.g., to make announcements, mention good news, or just say “hi”).
Each message was coded in one category only, even though some messages contained expressions of more than one type of uncertainty (e.g., a question about how to find a better specialist might also reflects expressions of concern about how to deal with public stares). In such cases, messages were coded based on “primary focus” (following Finn, 1999), and a choice was made for the type of uncertainty that seemed to be most prominently present, and most important to the general tone of the message.

Two coders (the author and a fellow student) coded the initial messages using the five-category coding system. The coders were trained by studying the rationale for the coding system and by thoroughly reading through numerous VBF posts on the VBF website to orient to the topic matter prior to coding. As a first step, the coders separately coded the same 25% of the sample to check for intercoder reliability. This provided a subsample of 36 messages. DIRC (Dozier Intercoder Reliability Calculator) was run on the intercoder data. Agreement on the uncertainty variable was 82.4%, with a Scott’s Pi of .89 (See Appendix A). Afterward, the coders compared data to see where most discrepancies occurred. Differences were resolved by discussion, after which the coders resumed independent coding practices.

TABLE 1. Classification of Expressed Uncertainty Types in Online VBF Messages
Uncertainty category Included topics in birthmark-related uncertainty
Complexity
what it is/diagnosis
what caused the condition
unexplained symptoms
Information quality(post-diagnostic)
whether a cosmetic problem is a real problem
how to positively influence the course of development of the condition
treatment effectiveness
treatment now or later
course of treatment
doctors’ expertise
what the condition means (consequences, implications)
who or what is right
which doctor to go see
how to pay for it
how to deal with insurance
Probability(predictability)
future development/potential for physical complications
expected pain associated with condition and/or treatment
whether a child will get teased or bullied because of the condition
potential for social acceptance/rejection because of the condition
potential for psycho-emotional trauma/damage
Integration(acceptance)
how to deal with the condition behaviorally/attitude toward child with birthmark
how to deal with the condition behaviorally/attitude toward environment
child’s attitude toward environment
how to explain the situation to child
how child will cope
how parent/environment will cope
expressions of need to find similar others
how to deal with emotions like embarrassment, anger, guilt, etc.
show or cover?
Non-uncertainty
announcements
informative topics

For the second part of the content analysis, the focus was on the responses to the initial message as the unit of analysis. The ensemble of responses to the initial message was coded per thread for overall dominance of type of support: informational support versus socioemotional support. This division was based on both (a) a two-category coding scheme used by Finn (1999), who discerned socioemotional helping mechanisms and task-oriented helping mechanisms in a content analysis of messages in an online self-help group, and (b) the coding system used by Lasker et al. (2005) that classified online messages in three categories: biomedical, socioemotional and organizational. The present study has kept the socioemotional label for those responses that predominantly emphasized matters such as feelings, encouragements, and a sense of community. The label “informational support” was created for those responses that emphasized the supply of medical information, (biomedical) advice based on comparative experiences, dealing with health insurance, etc. More qualitatively, the responses were also scanned for elements of authority and legitimacy following ideas from Galegher et al. (1998). In other words: when people ask for advice, what makes respondents trustworthy? What gives them expertise?

The type of content analysis applied to this study only gives a limited picture of a community and may miss important details such as specific demographics (e.g., age, time since diagnosis, insurance coverage), VBF visitors’ evaluation of their experiences, and more general levels of (unexpressed) birthmark-related uncertainty. A survey was developed to examine these elements (see Appendix B).

Survey

The initial aim of the online survey was to ask visitors of the VBF discussion forum more directly about their experiences and to measure overall levels of uncertainty. Also, since online messages tend to be viewed many more times than they are responded to (Galegher et al., 1998; Preece, Nonnecke, & Andrews, 2004), the survey would enable these “lurkers” to anonymously react to their experiences with vascular birthmarks and the VBF discussion forum, and thus be given voice without the need for public exposure on the board.

Respondents for the online, anonymous, self-report survey were recruited through a plea for participation that I posted in a personal message to the categories “Hemangioma Help,” “Port Wine Stain Help,” and “AVM Help” on the VBF community discussion forum (see Appendix C). My message provided a link to the survey, as well as a request for reactions and objections people may have to their messages being used for the study, and to the bigger issue of a research being done in their community. I explicitly mentioned in the research announcement that I am a community member myself as a parent of a daughter with PWS, who has previously posted to the VBF discussion forum and benefited from messages of others. I did this for several reasons: (a) to establish a connection and make people feel at ease with the research initiative; (b) to clarify my motivations to people, and show that my personal interest is also a community interest; (c) to motivate people to participate in the survey. They might be more inclined to do so knowing I am a community member, not just “some researcher.”

The questionnaire consisted of 51 questions, 7 of which were of a general demographical nature (e.g., age, education, gender). Another 7 asked about diagnosis and treatment (e.g., how long has it been since diagnosis? How many treatments have you had?), while 13 asked partially open-ended questions about experiences with support and information seeking both on and off the VBF board (e.g., what has been your most important source of information? Have you posted on the VBF discussion forum?).

The other 24 questions were part of two scales aimed at measuring level and type of illness-uncertainty. The first measure (question 21-32) is a short version of the 30-item Mishel Uncertainty in Illness Scale (MUIS) (Mishel, 1981, 1983). Item selection was done based on factor loadings higher than .50 from Mishel’s (1981) analysis. The other items were selected on applicability and appropriateness for this illness population (e.g., questions about extended hospital stays were left out; see also Mishel, 1983). The wording of some items was adapted to the specific vascular birthmark population, as well (e.g., the word “illness” was omitted, and generally substituted by “medical condition”). Previously, the MUIS has obtained reliability levels of .91 (Mishel, 1981) and .80 and .89 (Sheer & Cline, 1995). In this study, the MUIS had a Cronbach’s alpha of .90 after removing item 7 (“I understand everything explained to me”).

The second measure (question 33-44) was developed by the author and had not been tested prior to this study. The need for a second measure with additional items was motivated by the opinion that the MUIS is limited to those types of illness-related uncertainty that are dominated by the biomedical experience of diagnosis and treatment. Twelve more items were developed, loosely based on Babrow et al.’s (1998, 2000) meanings of uncertainty in illness and Sheer and Cline’s (1995) instrumentation to measure relational uncertainty in patient-physician interactions. This more biocultural measure included items such as “I am unsure what behavior is appropriate for this condition,” “I am unsure which questions are OK to ask the doctor,” and “Doctors don’t always have the necessary knowledge.” The new uncertainty scale had a Cronbach’s alpha of .85 after omission of item 3 (“It bothers me that I don’t know what caused this illness”).
All scale items were rated on a 5-point Likert-type scale (NO! – no – maybe – yes – YES!). Answers to the other survey questions were quantified and fit for statistical analysis where possible; answers to the open-ended questions were used to provide examples and to add elaboration of the motivations and characterizations of the respondents that quantitative analyses are unable to reveal.

Findings

During the three-month sample period, 659 messages were posted to 143 threads that were initiated by 132 different posters.

Initial Messages

Uncertainty was expressed in 86% (n = 123) of the 143 thread initiating messages. Of these 4 categories of uncertainty mentioned in the coding scheme, uncertainty related to information quality was expressed most often by far, in 64.2% (n = 79) of the messages (N = 123). Uncertainty related to complexity of the condition was dominantly present in 14.6% (n = 18) of the initial messages, whereas uncertainty about probability/predictability was more prominent in 13.8% (n = 17) of the initial messages. Only 7.3% (n = 9) of the messages emphasized difficulties with integration and acceptance. The dominance of informational uncertainty was often expressed in questions or remarks such as “Does anybody know how to deal with this because my doctor has no experience with this type of hemangioma,” “The pediatrician says a PWS will fade with time,” or “Can anyone help us better understand this growing thing? Does anyone have experience with internal hemangiomas of the esophagus?”

Of the 143 threads, 43.4% were initiated by obvious first-time posters (n = 62). A person was coded as a first-time poster based on explicit remarks such as “I am new here,” or “just found this website today.” Among the 62 thread initiators in the sample who obvious posted to the VBF discussion forum for the first time, fully 98.4% (n = 61) expressed uncertainty in their initial post. Only one newcomer did not express uncertainty, but merely offered her experience and support to those visitors who may need it. Among the 81 thread initiators who were not obvious newcomers (46.6%), expressions of uncertainty were present in 76.5% of their initial messages. Thread initiators who are also obvious first-time posters on the VBF discussion forum are significantly more likely to express uncertainty in their post than other thread initiators, c2(1, N = 143) = 9.71, p < .01. In other words: almost half the threads on the VBF discussion forum are initiated by newcomers, who almost always express uncertainty in these messages. A typical example: “Hi, I am new here and my 13-month old daughter was diagnosed yesterday with a hemangioma. The doctor insists it will go away on its own. Is this true???”

Among obvious first-timer thread initiators, uncertainty related to information quality was expressed even more often than among the other initiators: 64.5% (n = 40) of their messages expressed this type of uncertainty, versus 48.1% among the other initiators. Uncertainty about diagnosis (complexity) was also reported more often for these posters: this type of uncertainty was found in 17.7% of the initial posts, versus 8.6% for the others. Uncertainty about probability and integration were present in 9.7% and 6.5% respectively, versus 13.6% and 6.2%.
Most present among thread initiators on the VBF discussion forum are parents of children with a vascular birthmark. They made up 82.5% of the sample, whereas people who have a vascular birthmark themselves were responsible for the initiation of 13.3% of the threads. This seems influenced by the fact that the development of hemangiomas is most active in infants and young children, whereas PWSs and AVMs, although often visible at birth, generally continue as lifetime conditions. This became even more clear by using the cross tabulation procedure: of the 118 parent thread initiators in the sample, fully 83.1% (n = 98) posted in the category Hemangioma Help. In this category, only 1 thread was started by a person who had had a hemangioma herself. However, of the thread initiators who have a vascular birthmark themselves fully 73.7% (n = 14) posted on the PWS list, and 21.1% (n = 4) on the AVM list. Of all parent initiators, 11.9% (n = 14) posted in the PWS category, and 5.1% (n = 6) to the AVM category, c2(2, N = 137) = 50.02, p < .01. Fully 39.9% (n = 57) of the initial posts were from “new parents,” parents concerned about the birthmark of a child under the age of 18 months. The fact that the hemangioma category attracts many more messages than the PWS or AVM categories may be related to the fact that the latter conditions are much more uncommon types of vascular birthmarks than are hemangiomas.

Overall, many more women came on the discussion forum with questions than did men. Of all thread initiators, only 11.9% (n = 17) were obviously male, whereas 61.5% (n = 88) were obviously female, which is more than five times as many. Fully 92.1% (n = 82) of the parent thread initiators were female, and only 7.9% (n = 7) were male. However, among the selves with vascular birthmarks, threads were more often initiated by males, in 62.5% (n = 10) of the cases, versus 37.5% (n = 6) for females. These results were significant, c2(1, N = 105) = 29.83, p < .01.

Of all male parents, 100% (n = 7) posted in the hemangioma group, whereas this was only the case for 78% (n = 64) of the females. Of these, 14.6% (n = 12) and 7.3% (n = 6) posted in the PWS and AVM groups, respectively. Unfortunately, these findings were not significant, possibly due to the limited sample size. It is very well possible, however, that these numbers would show little change with a larger sample.

Some of the gender discrepancy among parents may be explained by the fact that the vast majority of vascular birthmarks become clear in the first few weeks after birth. This coincides with the period when traditionally, the majority of women are still at home with the baby, either on parental leave or as a stay-at-home mother. Parents who are at home with their child may be extra aware of blemishes on their babies’ skin and young mothers may already be extra worried induced by the hormonal fluctuations and big lifetime changes. However, previous studies have also found that women are more inclined in general to do health-related web searches than are men, especially when concerned with a specific disease or medical problem (Cotten & Gupta, 2004; Fox & Fallows, 2003).

A reason why, in this specific analysis, males unexpectedly dominated the “self” category can be found in the fact that one of the heaviest posters on the VBF discussion forum is male. In fact, in the sample period, he initiated a thread 6 times, mostly with non-uncertainty-expressing announcements and discussion openers. This obviously skewed the data, as there was only a total of 16 messages that were initiated by selves. The expectation is that, if we controlled for this person, the gender divide would be slightly dominated by women posters among the selves, too.

For types of experienced uncertainty that thread initiators express, no significant difference between the sexes was found. Both women and men most often expressed uncertainty related to information quality. A further dichotomization of the uncertainty variable into broader categories of informational uncertainty (complexity uncertainty and information quality uncertainty) and socioemotional uncertainty (probability and integration uncertainty) did not show any relation between gender and type of expressed uncertainty. Males and females dominantly expressed informational uncertainty in 81.8% (n = 9) and 81.1% (n = 60) of their uncertainty-related thread-initiating posts, respectively, and socioemotional worries in 18.9% (n = 14) and 18.2% (n = 2).

Responses

The total amount of responses in the sample was 817. During the three-month sample period, 516 new responses were posted to the 143 threads. The average number of responses per thread was 5.7. Almost three quarters of the messages (74.1%) received 6 responses or less. Only a handful of messages received up to 50 responses. Threads with more than 12 responses (14%) tended to have topics such as “Knowing that you’re choosing the right treatment approach,” “Will feelings of guilt ever stop?” “Is hormone treatment in AVMs always a no-no?” and “How to deal with questions from strangers.”

Some of these threads were as old as two years, but were still receiving replies in the sample period. The fact that people were posting to “old” threads in the sample period indicated that the topic in question was apparently still an issue they wished to discuss. By reading through all posts several times, it became clear that previous replies to an initial post are important to the tone-setting of the overall type of response. Once a first reply had chosen either a biomedical/informational direction as opposed to a socioemotional one, the tendency of the following replies was to continue on that note. Therefore, the decision was made to include all responses, including those dating before January 1, 2006, in the sample. Because of the large number of replies, time constraints, and because, as mentioned above, the tone of the responses hardly ever seemed to change once set by the initial response, general responses type was coded per response ensemble.
Two thirds of the responses (66.4%) were dominated by the supply of informational support, while 30.1% emphasized socioemotional support. Thread initiating messages that contained expressions of uncertainty related to complexity of the condition most often provoked informational responses: in 77.8% of the cases. Expressions of uncertainty related to information quality received informational responses 80.8% of the time. Unexpectedly, initial messages expressing uncertainty related to probability also more often received informational replies than socioemotional ones, in 82.4% of the cases. Uncertainty related to integration was predominantly responded to with socioemotional support, 75% of the time. These results were significant, c2(3, N = 121) = 13.06, p < .01.

Several reasons can be given to explain why discrepancies in the type of uncertainty expresses and the type of response given would occur. First of all, many initial posts express more than one type of uncertainty and respondents may, for whatever reason, not feel compelled to reply to the most dominant one. This may be based on what fits best with respondents’ own experience. Respondents may pick up on a type of uncertainty they consider to be more important, even though it is not dominantly expressed. For example, an initial message could state:

My daughter of 6 months has a very large hemangioma on her cheek. She has been on oral steroids, but we are unhappy with the advice the doctor is giving us. Should we be thinking about laser surgery? Can anyone tell me about their experience with general anesthesia? Your advice is greatly appreciated! I feel so bad and blame myself all the time. Will my daughter always have to live with her birthmark?

In this message, the uncertainty related to information quality is most dominantly expressed, but respondents may feel this new mother may be more helped by expressions of socioemotional support, e.g.,
You shouldn’t blame yourself. You did not put that birthmark on your daughter. There is a reason for everything, and this is God’s way of telling you that He wants you to go in a certain direction, not because He wants you or your daughter to be in pain.

Similarly, dominant expressions of socioemotional uncertainty may sometimes receive informational responses. An example of this is found in informational replies with invitations to join a support group on another website that is more specifically related to the described condition. Also, when people seem to be very distressed about their situation, often the informational advice of getting a second opinion or finding an expert is given in response.

The majority of threads with more than two replies included responses from heavy posters (determined as those who have posted to the VBF discussion forum more than 40 times). These heavy posters seem to be the people who have more authority and legitimacy on the board. Some of them have a specific place on the board, such as a VBF group moderator or a director of information services. The vast majority are ‘hands-on’ experts, people who rely on extensive experience with vascular birthmarks themselves (either as a parent or as a self with a vascular birthmark). Their tone is often humble (e.g., “I cannot tell you for sure because I am not a doctor but to me it looks like a port wine stain,” or “you may want to get a second opinion because what you have seems unlikely to be a hemangioma”), but in the sample their authority was never brought into question. Thread initiators very often responded to the responses with gratitude, voicing relief and the intent to follow up on good ideas (“e.g., Your messages have been so wonderful. Now I know I am not crazy!” “I finally made an appointment with a specialist! Thank you so much!!”) . Authority and legitimacy were also verbalized by reinforcement and explicit expressions of agreement with (often heavy) posters, e.g., “Sue-Ann has already said it all,” “As usual, I completely agree with James,” or “Betsy just gave you excellent advice.” Heavy posters are often spontaneously remembered as important people to the community on special days, such as days of important tests, treatments, or operations (e.g., “Please keep Paul in your prayers as he is getting surgery today”).

Survey Responses

My message to the VBF discussion forum announcing the research and asking for people to link to the survey was viewed 267 times. Over the course of the four weeks that the survey was open, a total of 30 people completed the survey. The average age of the 30 people who completed the survey the respondents was 35 (M = 34.66, SD = 9.15 years). The large majority (87%) of the respondents reported they were female (n = 26), while only 10% (n = 3) indicated they were male. One respondent did not reply to this question. Of the 26 respondents who reported their ethnicity, 92.3% were white (n = 24), while 3.8% reported being Asian (n = 1), and another 3.8% reported being of mixed ethnicity (n = 1). Fully 50% of the 28 respondents who reported their education level indicated they had a college degree (n = 14), and only 10.7% (n = 3) reported to not have gone to college. These data confirm findings from other studies that have repeatedly reported that people who use the internet for health-related purposes are predominantly female, white, and well-educated (e.g., Cotton & Gupta, 2004; Fox & Fallows, 2003).

Almost three quarters of the 30 respondents to the survey (73.3%) were parents of children with a vascular birthmarks (n = 22). Eleven respondents (26.7%) had a birthmark themselves. The majority of the parents (68.2%) who responded to the survey had a child with a hemangioma or hemangioma-related condition (n = 15). Six parents (27.3%) had a child with a PWS or PWS-related condition, and one parent (4.5%) had a child with an AVM or AVM-related condition. Of the 8 respondents who had a vascular birthmark themselves, 50% (n = 4) had a PWS or PWS-related condition, 37.5% (n = 3) had an AVM or AVM-related condition, and only 1 person reported to have a hemangioma or hemangioma-related condition (12.5%). Despite the small sample, these findings were significant: c2(2, N = 30) = 9.10, p = .01.

The vast majority of the parent respondents were female (95.2%). Only one father of a child with a vascular birthmark completed the survey. However, 2 of the 8 respondents who had a birthmark themselves were male (25%). Possibly due to the small sample, these findings were not significant: c2(1, N = 29) = 2.56, p = .18. At the same time, these findings do reflect the findings from the content analysis of the VBF discussion board messages that many more mothers are active online in relation to their child’s vascular birthmark condition than are fathers.
Of the 27 survey respondents who were willing to report about their health insurance status, 11.1% (n = 3) indicated they had no health insurance. Seventeen of 24 respondents (70.8%) who sought treatment for the vascular birthmark received full health insurance coverage. Another 12.5% (n = 3) received partial coverage (reimbursement up to 80% of the medical costs), and 16.7% of the respondents who sought treatment were denied coverage (n = 4). Of the 4 respondents who were denied coverage, 3 chose to have treatment anyway, despite the lack of financial assistance from their health insurance company. While these results are very telling for the sample, they cannot be considered generalizable, as they were not significant, most likely due to the small sample size, c2(2, N = 27) = 3.94, p = .14.

While respondents had only been able to link to the survey through the VBF discussion forum, a surprising 16.7% (n = 5) of the 30 respondents to the survey reported to never have posted to the VBF discussion forum, and were thus identified as “lurkers” (see Preece et al., 2004). Besides the fact that they had never posted to the VBF forum, no other obvious characteristics for this group were found. They were equally often parents and selves, newcomers and frequent return visitors to the VBF website, and had been diagnosed recently as well as a longer time ago. It remains unclear what compelled this group that has not participated in the online discussion, to complete the survey.

Based on the average scores on the Likert-type uncertainty scales, 51.7% (n = 16) of the respondent reported currently experiencing feelings of uncertainty related to the vascular birthmark (indicated by an average score of 3 or more on the ensemble of both scales). The fact that only a small majority of the respondents reported overall uncertainty may have to do with the wide variety in time lapsed since diagnosis and the phase of treatment they currently were in. In fact, fully 30% of the respondents (n = 9) reported that they had received the diagnosis more than 5 years ago, while only 26.7% (n = 8) reported to have received the diagnosis within the last year and a half. Interestingly, while the content analysis of the VBF messages showed that parents (particularly mothers) of newly diagnosed children are quite active as thread initiators on the discussion board (39.9%), they were apparently less inclined to participate in the survey. It is not unthinkable that overwhelming feelings of anxiety and confusion related to the diagnosis and current stage of the birthmark situation may have kept parents of more recently diagnosed children from participating in the survey, which in turn may also have influenced the lower than expected average uncertainty score. Uncertainty may often drive people with questions to the discussion board, yet it may make them unlikely candidates for a voluntary survey.

Both 11-item uncertainty scales showed to be very strongly and significantly related to each other (r = .78, p < .01). This provided the new, more bioculturally-oriented uncertainty scale with construct validity. An exploratory factor analysis was performed on the two uncertainty scales to see how, based on the data, the measures performed and whether the data would suggest a different, more comprehensive component structure reflecting different types of uncertainty. First, a principal component analysis was executed, guided by Eigenvalues greater than one and leveling in the scree plot. Oblique rotation was employed to improve interpretability of the factors.

The principal component factor analysis produced a Kaiser-Meyer-Olkin (KMO) coefficient of .62, and revealed six components with Eigenvalues greater than one. The scree plot, however, suggested leveling at the fifth component. Subsequent item extraction and rotation largely produced a four-factor structure that accounted for 67.3% of the common variance. The first factor loaded 6 “illness adjustment” items (a = .85), all from the second uncertainty measure (e.g., “I am unsure which questions are OK to ask the doctor,” and “I am unsure how to fit this condition into my life”). The second factor loaded 7 “illness predictability” items (a = .90), 6 of which originated from the MUIS (e.g., “I am unsure if the condition will get better or worse,” and “The symptoms continue to change unpredictably”) and one from the second uncertainty measure (“I am unsure when, which complications may happen”). The third factor loaded 5 “illness complexity” items (a = .80), 4 from the MUIS (e.g., “I have a lot of questions without answers,” and “It is difficult to know if the treatments or medications are helping”), and one from the second uncertainty measure (“The complexity of this condition makes it hard for me to understand it”). The fourth factor loaded 2 “lack of medical knowledge” items (a = .72), both from the second uncertainty scale (“I am unsure what behavior is appropriate for this condition’” and “Doctors don’t always have the necessary knowledge”).

The found factors largely reflect a division in types of uncertainty similar to the content analysis coding scheme and Babrow’s (1998, 2000) meanings of uncertainty. These findings imply that the illness uncertainty experience takes place on several levels. People affected by a vascular birthmark may not only experience uncertainty related to symptoms and treatment, but also related to integration, predictability, and life adjustment. However, all factors, especially the first three, were still highly correlated (see Table 2), indicating the division in explanation between the factors is not as distinct as would be desirable when creating new, separate, meaning-specific scales. In other words, the factors may still be measuring largely the same concept, and with another sample, at another time, they may come out differently. Interestingly, though, the fourth “lack of medical knowledge” factor seems to take a more independent position. When looking again at the average scores on the scales (see last column in Table 2), we see that this scale received an average score of 4.24 (on a range of 1 to 5), expressing a very high level of respondent agreement with the two scale items.

TABLE 2. Uncertainty correlations and average scores after factor analysis
1 2 3 Average
Illness Adjustment 2.25
Illness predictability .59** 2.61
Illness complexity .70** .78** 2.50
Lack of medical knowledge .26 .32 .25 4.24

Note. N = 29.
**p < .01.

The high score on this scale also reflects a repeatedly reported problem the vascular birthmark community struggles with: the fact that, in general, doctors know little about vascular birthmarks and that a lot about vascular birthmarks has yet to be discovered. This sentiment is also voiced by the answers to the survey question “What is the most important thing you found on the VBF website?” Fully 43.3% answered “information”, and 23.3% reported a mixture of information and support, whereas only 33.3% answered “support” only. Fully 76% of the respondents answered that their most important source of information had been found on the internet, while 17.2% indicated their most important information had come from both online and offline sources equally, and only 6.9% had found their most important source of information offline (e.g., by a hospital or doctor).

As most important source of information in dealing with their vascular birthmark situation, the VBF was mentioned by 30% of the respondents. As one respondent described: “Based on others’ experiences as posted, I realized that we should seek a second opinion.” Another respondent reported that, based on the VBF discussion board messages, she realized that “our daughter should have had treatments a long time ago…the doctor I trusted knew nothing.” Others reported their most important gain was “learning we were not receiving needed treatment.”

A slight minority, 46.4% of the respondents, mentioned the internet as a primary source of support. Another 14.3% claimed to have found equal support online and offline, whereas 39.3% most importantly relied on offline sources for support (e.g., family, religion). Messages on the VBF discussion board had made 34.5% of the respondents at times feel more confused, and thus more uncertain. However, survey results also indicated that messages on the VBF discussion board had made fully 100% of the respondents feel better. “Just knowing that there are people out there who understand and don’t think you are a worrywart,” signified a turning point for one of the respondents, while others reported that VBF messages helped them with “accepting that my feelings were normal and that other people feel the same.”

Conclusion

Unsurprisingly, findings from the content analysis of the VBF discussion forum messages and the survey show that the experience of uncertainty in the vascular birthmark community is dominated by a lack of (medical) information supply for this community. This may partially be due to the fact that many doctors tend to be ignorant on the topic of vascular birthmarks because they have not been well informed themselves (see Shannon & Marshall, 1997), and partially because a lot about vascular birthmarks is still unknown (e.g., its cause, and how to treat people with birthmarks to make them “look normal”) and because of its unpredictable nature by definition (“no two birthmarks are the same”).

Expressions of uncertainty were more dominantly present on the VBF discussion forum than in the survey responses. A reason for this may be that when uncertainty goes up, the inclination and willingness to participate in surveys goes down. Also, the survey respondents seemed to feel more confident overall, maybe because, on average, they had known about their diagnosis for a longer period of time. However, survey respondents readily acknowledged the painful absence of adequate information for people who are confronted with a vascular birthmark diagnosis, and the difficulty and the severe consequences this may entail for getting adequate treatment.

Remarkably, many more women are active in the VBF community than are men. On the VBF discussion forum, posters were more than five times more likely to be female than male, and more than eight times more women participated in the survey than did men. While it is a known fact that women show more health-related internet activity than do men (see Cotton and Gupta, 2004; Fox & Fallows, 2003) the general discrepancy between the sexes in internet health matters is not nearly as prominent as it is in the VBF community. A reason for the high number of women on the VBF discussion board may be that many of the VBF members are parents of young children with vascular birthmarks, and that in families in general, the responsibility for the (health) care of children still turns out to be a mother’s responsibility.

People posting on the VBF discussion forum are very often either newcomers or heavy posters. The VBF discussion forum is not so much a vibrant, tight community where established internet friends meet on a regular basis to discuss their ongoing birthmark-related whereabouts. For these types of more sustained support-oriented activity, new posters on the VBF board are often referred to other websites that feature listservs and support groups. The VBF discussion forum appears to be more of a stage of entrance, a low threshold meeting place where many people who have just recently initiated a search on vascular birthmarks make their first steps into the vascular birthmark community. Return visitors and heavy posters often indicate they also have memberships in other, more condition-specific birthmark-related support groups. As such, the VBF discussion forum fulfills an important role as a warm, welcoming starting place for people who are trying to find their way in the world of vascular birthmarks. Heavy posters on the VBF discussion forum are ready to help these new people become acquainted with the vascular birthmark community and to supply them with information and support. In fact, the opportunity to help others was often mentioned by survey respondents as an important reason to keep returning to the VBF discussion forum.
People with vascular birthmarks almost all face the challenge of living with a stigmatizing condition. This is not only the case for those with lifetime conditions like PWS and AVM, but also for those marked by not yet involuted hemangiomas or by scars of improperly treated birthmarks. Facing psychosocial consequences and potential social rejection is a highly uncertainty-provoking aspect of dealing with illness (Brashers at al., 2002). However, matters of integration and acceptance were less often discussed on the VBF discussion forum than expected (see Davison, et al., 2000; Galegher et al., 1998; Muthaseb & Wright; Reeves, 2000). A reason for this may again be found in the idea that the VBF discussion forum is a place of initiation, dominated by newcomers with acute questions that are most often of a biomedical, information-related nature. It is very well possible that the discussion in other, more birthmark type-specific online support groups encompasses a much larger proportion of topics related to dealing with stigma and acceptance.

The classification of informational uncertainty as one category grouped together a wide variety of issues. Therefore, the quantitative results don’t reflect well how often important topics such as getting adequate coverage from health insurance companies were discussed. The survey, however, revealed a practice that has been alerted to by McClean and Hanke (1997): an important fraction of the respondents indicated to have been denied complete or some coverage for expensive medical procedures such a laser therapy and surgery to treat vascular birthmarks, possibly because these treatments were considered to be purely “cosmetic.” While vascular birthmark experts today, without exception, recognize the importance of treatment in many cases (e.g., Troilius et al., 1998; Schiffner et al., 2002), some insurance companies apparently still manage to escape financial responsibility and force people to turn to alternative ways of financial support.
Future research could look more closely at matters such as insurance companies’ decision making processes in requests for coverage for the treatment of vascular birthmarks. Also, a further exploration of patterns in discussion on other birthmark-related internet support groups should reveal if indeed discrepancies exist with the topics discussed on the VBF discussion forum. Based on the findings of this study, it is recommended that future studies of types on expressed uncertainty (in online messages) further specify uncertainty categories. Many of the sampled VBF messages ended up in the “information quality” category. While the number of categories was deliberately kept low to be able to get significant results, a further specification of types of information people look for (e.g., biomedical, insurance-related) would provide more comprehensive picture of VBF visitors’ most prominent informational needs.

Lastly, future research aimed at the vascular birthmark community could more explicitly reveal a link between the negligence in education of medical professionals and its consequences for vascular birthmark-affected families. Findings of such a study may not only be beneficial to the vascular birthmark community, but also to other health-related communities centered around complex, rare disorders (see Patsos, 2001).

References
Ainbinder, J. G., Blanchard, L. W., Singer, G. H. S., Sullivan, M. E., Powers, L. K., Marquis, J. G., & Santelli, B. (1998). A qualitative study of Parent to Parent support for parents of children with special needs. Journal of Pediatric Psychology, 23, 99-109.
Babrow, A. S., Hines, S. C., & Kasch, C. R. (2000). Managing uncertainty in illness explanation: An application of problematic integration theory. In B. B. Whaley (Ed.), Explaining illness: Research, theory, and strategies (pp. 41-67). Hillsdale, NJ: Erlbaum.
Babrow, A. S., Kasch, C. R, &. Ford, L. A. (1998). The many meanings of “uncertainty” in illness: Toward a systematic accounting. Health Communication, 10, 1-24.
Babrow, A. S., &. Kline, K. N. (2000). From “reducing” to “coping with” uncertainty: Reconceptualizing the central challenge in breast self-exams. Social Science & Medicine, 51, 1805-1816.
Brashers, D. E., Goldsmith, D. J., & Hsieh, E. (2002). Information seeking and avoiding in health contexts. Human Communication Research, 28, 258-271.
Brashers, D. E., Neidig, J. L., & Goldsmith, D. J. (2004). Social support and the management of uncertainty for people living with HIV or AIDS. Health Communication, 16, 305-331.
Burgoon, M., & Hall, J. R. (1994). Myths as health belief systems: The language of salves, sorcery, and science. Health Communication, 6, 97-115.
Calhoun, R. D. (1998). Practice guidelines for patients with rare cancers. Cancer Practice, 6, 247-250.
Cohen, M. H. (1993). The unknown and the unknowable – Managing sustained uncertainty. Western Journal of Nursing Research, 15, 77-96.
Cotten, S. R., & Gupta, S. S. (2004). Characteristics of online and offline health information seekers and factors that discriminate between them. Social Science & Medicine, 59, 1795-1806.
Davidow, J. (2003). Marked for life: A memoir. New York: Harmony.
Davison, K. P., Pennebaker, J. W., & Dickerson, S. S. (2000). Who talks? The social psychology of illness support groups. American Psychologist, 55, 205-217.
Dickerson, S. S., Boehmke, M., Ogle, C., Brown, J. K. (2006). Seeking and managing hope: Patients’ experiences using the internet for cancer care. Oncology Nursing Forum, 33. Retrieved March 6, 2006, from http://www.ons.org/publications/
journals/ONF/volume33/issue1/330138.asp
Dutta-Bergman, M. J. (2005). Developing a profile of consumer intention to seek out additional information beyond a doctor: The role of communicative and motivation variables. Health Communication, 17, 1-16.
Finn, J. (1999). An exploration of helping processes in an online self-help group focusing on issues of disability. Health & Social Work, 24, 220-231.
Ford, L. A., Babrow, A. S., Stohl, S. (1996). Social support messages and the management of uncertainty in the experience of breast cancer: An application of problematic integration theory. Communication Monographs, 63, 189-207.
Fox, S., & Fallows, D. (2003). Internet health resources: Health searches and email have become more commonplace, but there is room for improvement in searches and overall Internet access. Retrieved May 13, 2006, from Pew Internet and American Life Project Web site: http://www.pewinternet.org/report_display
.asp?r=95
Galegher, J., Sproull, L., & Kiesler, S. (1998). Legitimacy, authority, and community in electronic support groups. Written Communication, 15, 493-530.
Han, H. R., & Belcher, A. E. (2001). Computer-mediated support group use among parents of children with cancer – An Exploratory study. Computers in Nursing, 19, 27-33.
Hines, S. C. (2001). Coping with uncertainties in advance care planning. Journal of Communication, 51, 498-513.
Klemm, P., & Hardie, T. (2002). Depression in internet and face-to-face cancer support groups: A pilot study. Oncology Nursing Forum, 29. Retrieved March 6, 2006, from http://www.ons.org/publications/journals/ONF/Volume29/Issue4/
290445.asp
Lasker, J. N., Sogolow, E. D., & Sharim, R. R. (2005, March 26). The role of an online community for people with a rare disease: Content analysis of messages posted on a primary biliary cirrhosis mailinglist. Journal of Medical Internet Research, 7, Article e10. Retrieved February 6, 2006, from http://www.jmir.org/2005/1/e10/
Macias, M., Lewis, L. S., & Smith, T. L. (2005). Health-related message boards/chat rooms on the web: discussion content and implications for pharmaceutical sponsorships. Journal of Health Communication, 10, 209-223.
McClean, K., & Hanke, W. C. (1997). The medical necessity for treatment of port wine stains. Dermatologic Surgery, 23, 663-667.
Mishel, M. H. (1981). The measurement of uncertainty in illness. Nursing Research, 30, 258-263.
Mishel, M. H. (1983). Adjusting the fit: Development of uncertainty scales for specific clinical populations. Western Journal of Nursing Research, 5, 355-369.
Mishel, M. H. (1988). Uncertainty in illness. Journal of Nursing Research, 20, 225-232.
Mishel, M. H. (1990). Reconceptualization of the uncertainty in illness theory. Journal of Nursing Scholarship, 22, 256-262.
Muthaseb, A., & Wright, K. (2005, May). Perceptions of support group advantages, types of support, and support providers within online support groups dealing with health-related issues. Paper presented at the annual meeting of the International Communication Association, New York, NY.
National Organization of Vascular Anomalies (2004). Hemangioma & Vascular Malformation Information and Assistance. Retrieved April 9, 2006, from www.novanews.org
Patsos, M. (2001). The internet and medicine: Building a community for patients with rare diseases. Journal of the American Medical Association, 285, 805.
Preece, J., Nonnecke, B., & Andrews, D. (2004). The top five reasons for lurking: Improving community experiences for everyone. Computers in Human Behavior, 20, 201-223.
Reeves, P. M. (2000). Coping in cyberspace: The impact of internet use on the ability of HIV-positive individuals to deal with their illness. Journal of Health Communication, 5, 47-59.
Schiffner, R., Brunnberg, S., Hohenleutner, U., Stolz, W., & Landthaler, M. (2002). Willingness to pay and time trade-off: Useful utility indicators for the assessment of quality of life and patient satisfaction in patients with port wine stains. British Journal of Dermatology, 146, 440-447.
Shannon, L. R., & Marshall, C. (with Waner, M.). (1997). Birthmarks: A guide to hemangiomas and vascular malformations. Minden, NV: Women’s Health Publishing.
Sheer, V. C., & Cline, R. J. (1995). Testing a model of perceived information adequacy and uncertainty reduction in physician-patient interactions. Journal of Applied Communication Research, 23, 44-59.
Smith, M. L., & Sinson G. P. (2006, April 4). Intracranial Arteriovenous Malformation. Retrieved April 9, 2006, from http://www.emedicine.com/med/topic3469.htm
Syed, S. (1999). Vascular birthmarks: Update on presentation and management. Current Paediatrics, 9, 20-26.
Troilius, A., Wrangsjö, B., Ljunggren, B. (1998). Potential psychological benefits from early treatment of port-wine in children. British Journal of Dermatology, 139, 59-65.
Troilius, A., Wrangsjö, B., Ljunggren, B. (2000). Patients with port-wine stains and their psychosocial reactions after photothermolytic treatment. Dermatologic Surgery, 26, 190-196.
Vascular Birthmark Foundation. (n.d.a). Birthmarks. Retrieved February 18, 2006, from www.birthmark.org
Vascular Birthmark Foundation. (n.d.b). History of the VBF. Retrieved February 8, 2006, from www.birthmark.org
Vascular Birthmark Foundation. (n.d.c). Vascular Birthmark Foundation. Retrieved February 5, 2006, from www.birthmark.org