There are two main ways of treating a broken ACL. The ligament can be reconstructed by surgery, or surgery can be avoided and instead, the patient can follow an intense program of physical therapy to strengthen surrounding muscles to compensate for the missing ligament. Surgery tends to be the favored option by athletic people because, for most people, it brings nearly full recovery and return to sporting activities. However, recovery may take a year and be painful.

The choice of treatment is, however, not as straightforward as it might seem. If surgery is chosen, patients must select between different versions of the reconstruction. Furthermore, patients need to remember that people respond differently to treatments, have different tolerances to pain, and experience varying recovery times, so what works for one person may not for another. Age, gender, body weight, type of life-style and attitude also vary from person to person and influence the success of the surgery. Of course, the skill of the physician is an important variable too. Having made a choice, the patient then has to work through the recovery phases, which may be tedious and painful. In addition there may be stresses associated with changes of life-style after surgery and the uncertainty of whether full recovery will be achieved.

Orthopedic surgeons can answer medical questions and may be able to draw on observations of hundreds of patients, but what most patients want to know is, what it was really like for other patients. They want to hear first-hand accounts of their experiences. How much did it hurt? How does what I am experiencing compare with what you experienced? If I do this, what should I expect? One way of getting answers to such questions is to seek other patients on web bulletin boards, of which, there are now several thousand. Some are produced as suites of boards by the same originator(s), and some, like Bob’s bulletin board, are the work of caring individuals who want to share their experience and help others.

The ACL bulletin board, known as Bob’s bulletin board after its originator, forms part of an ACL web site. The bulletin board (http://www.cnct.com/~bwillmot/knees/wwwboard/) started in April 1996. The number of postings per day increased from around twenty-four in May 1996, to over one hundred in April 1997. By October 1997 there was an archive of around nine thousand messages. Messages are threaded so that the headers of responses are indented and positioned below the header of the message to which the reply applies. The bulletin board is rudimentary. Messages cannot be ordered and searched by date, topic or sender’s name. Some people use login names, some use first names, some full names, some just give initials and some give a description. There are no guidelines on how to specify topic lines, so styles vary and some are more helpful than others. Examples of typical topic lines are:

Help … Tomorrow is D-Day! (reference to surgery).

I don’t know what to do!!! (reference to just discovering the extent of his injury)

Man, I AM NERVOUS! (concern not only about having surgery the next day, but also about whether he will be able to play football again)

Although, guesses can be made about what follows, the topic(s) being discussed are not obvious.

The gender ratio is close to two males to one female and is similar to that reported by Pitkow and Kehoe, (1997) for the general population of Internet users. In 1996 when the board started there was a higher proportion of men. In a pilot study in March 1997 the gender ratio was 4 males to one female. Most posters are between twenty and forty years of age, and share a strong love of sports. This provides a clear focus for the group and common understanding between them, which undoubtedly contributes to the success of the community (Whittaker et al., 1997). The scope of most discussions tends to be narrow and clearly focused. The bulletin board is open to anyone, so the occurrence of acrimonious posts (i.e. flaming), spamming and posting junk mail, would not be surprising but such behavior is unusual. Bob lightly reviews messages, but he rarely intervenes.

Part of the membership of the ACL community appears fairly transient. Most of the new people on the board have recently sustained the injury, but there are a few old timers who keep coming back. There have been a few postings from people professionally involved in orthopedic treatment, such as a nurse and a specialist in knee braces, but there is no sign of physicians. The tone of the messages suggests a sense of equality and trust within the group. Communication between members is helpful, empathic, often altruistic and sometimes cathartic. People want to share their experiences, learn from others and support each other.

A content analysis of 500 archived messages, from the ACL bulletin board, sampled in batches of 100 at approximately two monthly intervals, was analyzed (Preece, 1998). 312 messages were posted by men, 156 by women and 32 could not be classified for gender. 251 different people posted these messages. The largest number posted by one person was 24. These were messages posted by Bob and most were to draw attention to previous posts on a topic of interest. For example, common topics that new-comers ask about include: is surgery necessary, what is the difference between different surgical procedures, how long is recovery and so on. At the time of this study additional useful information, including a video of arthroscopic surgery, was available on the web site but frequently asked questions were not included. 150 messages were single posts by one person. Figure 1 contains a summary of these findings. A detailed discussion of the analysis and the methodology used in this study is contained in Preece (1998).

Figure 1 Messages sent per individual in the ACL bulletin board study (Excel chart)

The analysis shows that 76.8% of the posts were empathic to some extent. Some directly asked or offered support, some simply told their story, many asked or answered questions by drawing upon their own personal context. Side-bar A shows examples of some of the different ways of communicating empathy.

Only 17.4% of the postings were purely factual questions or answers, and showed no empathy or reference to personal experience, as illustrated by the examples in side-bar B. 5.8% fell into the category of ‘other’ and most of these were jokes. Occasionally there are signs of frustration, when new-comers, anxious to find out more about their recently suffered injury, ask basic questions that have been discussed many times before. Usually someone politely refers the inquirer back to an earlier discussion. The community is surprisingly tolerant and annoyance is rarely shown.

In the sequence of postings that follow, Debbie directly asked for support and Ken, Kath and Mike responded by empathizing with her situation, and suggesting ways to cope. Notice also the expression of warmth and encouragement to ‘hang in there’, an invitation to come back and tell how she is progressing, a note that she is welcome to contact the other ‘veggies’ at any time and wishes of good luck. Considering these people only know each other through this bulletin board, the warmth, caring and empathy conveyed in their postings is remarkably strong.

The next three comments show empathy about life-style. Dave is frustrated that he can’t ride his bike in the beautiful weather. Brad and Bill respond with comments about their own situations, which they point out are worse but in a friendly, joking manner which is supportive.

Some people use the board to tell their story and as a way of inviting others to contact them. The story below from Jimmy is an example of this. He is not asking questions, he tells his story and presumes that when he comes back with more details about his physical therapy, people on the board will be interested in listening. He also wants to encourage and support others.

Jimmy’s story is typical of an indirect request for empathy. He does not ask for support or help directly but instead tells his story which often has the effect of eliciting empathic responses. In our study we observed that men adopted this approach more frequently than women, who made more explicit requests for empathy. The men in this study also asked more factual questions than women (Preece, 1998a).

There were fewer people who asked straight factual questions or who gave answers in which there was little or no reference to their own personal situation, such as the comments from Gerry and Rob.

Not surprisingly, patients’ needs change between the time of their injury though to their recovery. The kinds of questions asked changes and so does their need for empathy. Furthermore, the balance between needing information and empathy varies too. The patients’ messages suggest that three main stages can be identified along the path from injury to recovery: from injury to diagnosis, from diagnosis to treatment, from treatment to recovery, as shown in figure 2.

Some people get their injury diagnosed very quickly, others need to visit two or more doctors and may seek the confirmation of an MRI. At this first stage, the factual content of messages is about details of diagnosis and the kinds of treatments that are available. The empathic comments generally consist of people telling their stories, frustrations about not being able to play sports, and concerns about how long it will be before they can play again.

The time between diagnosis and deciding upon treatment is when the most detailed factual questions are asked. Patients want to find out about different treatments, check their understanding of the procedures involved, get estimates of the likely amount of pain and discomfort, and check the prognoses for recovery. They also want to hear from others who are further along the path to recovery. Patients also get frustrated and scared and seek the support of others, but the need for factual information seems to dominate at this stage.

After treatment, which for an ACL injury generally means some form of surgery, patients are eager to hear how their experiences compare with those of others. This is the time when empathy is strongest, especially for patients who experience pain, who are not sleeping well and who wonder how long they will feel like this. The lucky ones want to check their progress with others who have gone before them. Some people count the days until they can resume their favorite sports … biking, football, skiing etc. and want information about the best physical therapy workouts to get them back playing again.

People vary enormously but, in general, the time before treatment is when most fact finding occurs, with the time just after diagnosis being the most intense of all. The same questions frequently get asked again and again. Empathic messages are sent at all stages, but the need for empathy is often strongest just after surgery.

Figure 2 Injury-diagnosis-treatment-recovery path

The messages sent to Bob’s bulletin board illustrate the important role of empathy in this community. People want to make contact with others, tell their stories and be heard, offer and receive support, as well as get factual information about their problem. Learning the medical facts is only part of the communication on this board. Even those who ask or offer factual content usually do so in the context of their own injury, recovery and broader life-style. These messages are from patients concerned about their condition and it is common for them to talk about their own experience. Messages from other patient support groups, also show a high level of empathy (Preece and Ghozati, 1998). Women’s health groups are often the most strongly empathic, particularly those which have joining requirements, rules and are moderated.

Research in psychotherapy (Etchegoyen, 1991; Ickes, 1997) informs us that empathy is present in most communication, but in many work situations it may be subtle. Much of the research in computer supported co-operative work (CSCW) and computer mediated communication (CMC) focuses on performance in formal tasks, where empathy is usually not obvious. The dramatic increase in empathic communities provides exciting opportunities for researchers to take a new perspective, which acknowledges the role of empathy in informal online communication.

The ACL bulletin board, like many bulletin boards, has little structure and is an example of the ad hoc evolution of bulletin board design (Gaines et al, 1997), which is just beginning to change as designers become more conscious of communities’ needs. Basic tailoring facilities and emoticons are examples of change. As the number of people online increases, users will request better usability. For example, they will question why they cannot do seemingly straightforward things, like easily search previous messages and archives. Some ACL patients wanted to contact others similar to themselves. Another issue discussed, was the frustration experienced when newcomers ask questions that have already been discussed at length. In addition, ways of dealing with well-known problems, like flaming, continue to be important for all communities.

Figure 3 shows some key design requirements for empathic bulletin board communities. At first glance these issues appear like those needed by any online community. Empathic communities are at the extreme end of the empathic/non-empathic spectrum of communities. So the difference lies in the amount of emphasis given to supporting empathy. In order to support the high level of empathy that makes empathic communities function, their requirements need to be met in ways that do not threaten empathic communication. There may, therefore, be times when efficiency has to be balanced against supporting empathy and trade-offs made. For example, FAQs are efficient but may effect empathy among patients, as discussed below. Similarly, a balanced approach is needed to developing techniques for information searching and other key user activities (center of diagram). Ways of dealing with concerns like privacy (left of diagram) may require more human involvement than in other communities. People in distress also tend to be vulnerable, so guarding against misleading information (particularly medical information), imposters and scams is a serious issue. Good resources on the web (top right of diagram) are also needed with ways of accessing them from within a bulletin board (right of diagram). For example, it would be helpful to click on a term in a message and call up a web browser. Tailoring the bulletin board display and keeping personal files is useful too.

Figure 3 Design for empathic bulletin board communities

Most people join Bob’s bulletin board soon after their ACL injury. They want to find out as much as possible about the various treatment options available. Sometimes, people also want to find others with whom they can closely identify in terms of gender, age, life-style, sporting preferences, level of sporting activity, and body mass to see what choices they made and how successful they were. For example, a patient might want to find women over forty years of age, weighing around 120 pounds, who are moderately active. The patient might want to gauge where the women are in the recovery process, to review their messages and others in the same thread or on related topics. At present the only way to do this is to scan the list of names for those that look female or post asking people who fit the description to identify themselves. Having found potential candidates, the next stage might be to contact them directly.

Providing this functionality raises privacy issues as well as technical ones. People may not want to receive email from strangers, so making this easy would not be desirable. Other solutions need to be found. For example, people could be invited to complete a simple on-line form, which would be active for a limited period of time. Then there would be the opportunity to renew or discard the form. There could be a facilitator who would make sure that requests for contacts are honorable and agreeable to the person being asked. The facilitator would then initiate an introduction..

Information visualization techniques, like those used in the FilmFinder (Ahlberg and Shneiderman, 1994), could enable participants to visualize data from community members collected via forms. Using such techniques, it would be possible to visualize the relationships between combinations of variables such as age, gender, time since the accident, time since surgery and so on. Searching strategies can be adapted on the basis of visual feedback to enable users to hone in on specific individuals.

Research on recommender systems may also be useful in this context. Recommender systems are like collaborative filtering, but the term is broader and includes systems that may not explicitly collaborate with recipients (Resnick and Varian, 1997). Some of these systems enable users to enter their preferences directly, in addition to the system tracking choices from past user behavior. Using such systems, it might be possible to alert members of communities to posts that might be of interest to them. Users could then decide whether or not they want to follow-up on the recommendations. GroupLens (Konstan et al., 1997), for example, performs a similar service for Newsnet readers. Studies of GroupLens indicate that that the service is useful. However, it was difficult to get adequate ratings upon which to base future searches because of the small number of articles read by each Newsnet user. The narrow focus of many empathic bulletin board communities may be a successful avenue for exploring such systems.

One problem with these suggestions is that they have implications for individuals’ privacy, which must be protected. A second is that they involve substantial software development. In the short term improved ways of tailoring bulletin board displays and book-marking, would improve usability. Some agreed formats for message headers (i.e. topic lines and user names) would also help.

When I joined the ACL community I wondered why there were no FAQs on the associated web pages. Later, as I used the board more, I realized that, as with many obvious solutions, there could be a serious problem with this one. Providing concisely written answers in FAQs may inhibit people from asking questions in case they are thought unnecessary and a nuisance by others. FAQs might encourage intolerance and abruptness. Comments like: ‘you should check the FAQs first before posting’ or ‘don’t waste people’s time by asking questions that are already answered in the FAQs’ could reduce or even eliminate empathy and change the nature of the community. Since I did the study, FAQs have been added to the web site. So far, my concerns about destroying empathic communication have not been born out and I hope they will not.

A balanced approach to supporting efficient information exchange and empathy is needed and the solutions may vary from one community to another depending on the nature of the community, the topics of discussion, the structure and usability of the supporting software. For example, Ackerman and Palen (1996) observed a different solution in their study of the Zephyr system, which is a conferencing system used by MIT students to discuss programming problems. They concluded that there was no need for a system memory because the users did not mind answering the same questions again and again. Of course, discussions of technical issues are likely to be less empathic than personal health. In a study of the general use of email in a professional setting, Whittaker and Sidner (1996) point out that many email conversations are multi-threaded with several conversations going on at once, information getting lost and the same questions being asked again and again. The authors do not mention if this is a problem.

Many concerns for empathic communities can be restrained by unobtrusive moderation of some form, rules and by-laws. Several roles have been identified for moderators of online communities (e.g. Berge, 1992) in addition to stopping flaming and other anti-social behavior. Some moderators lead discussions and keep them on track, but this is less important for empathic community than for many other groups. Moderators may also act as experts, answering questions or directing people to previous messages or resources, as Bob does for the ACL group. Moderators of empathic communities may need to take more diverse roles, channeling discussion of different topics, maybe encouraging people to seek expert advise, keeping watch for scams and so on. In particular, as groups get very large, moderators will need to create meaningful sub-groups for discussion of specific topics. Deciding when to do this could be a sensitive issue for empathic communication. It is well known that moderators change the nature of online communication (Whittaker, 1996; Whittaker and Sider, 1996; Konstan, 1997), and supporting empathy requires skill. Mediators are becoming more widespread. Unlike moderators, mediators only take action when called to do so by the community.

As well as ensuring that an empathic climate prevails, moderators need to look out for people posing as medical experts or counselors. It is easy to imagine instances of people posing as professional medics in order to give or, worse still, sell advice. There has been no evidence of this behavior on the Bob’s bulletin board, but it could be a concern. Incidents of people posing as someone else (Rheingold, 1993) and gender swapping have been well-documented (Sproull and Kiesler, 1991, Turkle, 1995). Imposters selling expensive homeopathic medicine, promising miracle cures, were observed in several empathic groups. Warnings spread through the groups very quickly but some people made purchases before the scam was exposed. Awareness that people are watching for scams may deter them, but the problem is likely to increase as more people come online.

Joining requirements may also become more widespread to discourage those without bona fide interest. More public explanations of moderators roles and clear descriptions of community rules might help discourage unacceptable behavior and support empathy. Further research is needed into the details of these issues in relation to empathic communication.

Private communication channels are useful for contentious one-to-one debate, for discussing topics of narrow interest, or a personal nature (Ackerman and Palen, 1996). Furthermore, there are limits to the amount and kind of information that a person will make available to the public at large, so private forums for communication are needed (Kautz et al., 1997). One-to-one communication is also known to be liked by women and girls (Tannen, 1995). Facilitating private communication is likely to encourage empathy; particularly when the topics of discussion are personal.

The social ecology of human groups is delicate. Just as in ecosystems of animals and plants, the dynamics of human social ecology is such that many variables are inter-related. Change one and there can be a ripple of change through the whole system in unexpected ways. The dynamics of on-line communities that have only text as a medium for expression may be even more sensitive, because they lack so many of the feedback cues that we are used to in face-to-face communication. Even quite small changes to the physical (i.e. software) and social structure of the community can have a big impact on the way it functions.

Communities are forming on the web at a phenomenal rate for all kinds of purposes (Anderson et al., 1995; Pitkow and Kehoe, 1997). By better understanding the role of empathy in a cross-section of these communities, we can investigate and design tools to support empathic communication as well as factual information exchange. Well-designed empathic communities will improve the quality of life for thousands of people.

I wish to thank Bob Wilmot for answering my questions and for developing Bob’s bulletin board, which is a source of information and comfort to thousands of knee sufferers. Kambiz Ghozati did most of the data analysis for this study and provided many useful insights. I also thank the reviewers and editorial staff of Interactions Magazine and Ben Shneiderman for their comments.

Ackerman, M. S. and Palen, L. (1996) The Zephyr help instance: Promoting ongoing activity in a CSCW system, Proceedings of ACM CHI’96 Conference, 268-275.

Ahlberg, C. and Shneiderman, B. (1994) Visual information seeking: Tight coupling of dynamic query filters with starfields displays, Proceedings of ACM CHI’94 Conference, 313-317.

Anderson, R. H., Bikson, T. K., Law, S. A. and Mitchell, B. M. (1995) Universal Access to E-mail: Feasibility and Societal Implications, RAND: Santa Monica, CA.

Berge, Z. L. (1992) The role of the moderator in a Scholarly Discussion Group (SDG) http://star.ucc.nau.edu/~mauri/moderate/zlbmod.html

Carroll, J. M. and Rosson, M. B. (1996) Developing the Blacksberg Village. Communications of the ACM, 39, 11, 69-74.

Cohill, A. (1997) Ed. Community Networks: Lessons from Blacksberg, Virginia.

Etchegoyen, R. H. (1991) The Fundamentals of Psychoanalytic Technique, London and New York: Karnac Books.

Gaines, B. R., Chen, L-J. L., and Shaw, M. L. G. (1997) Modeling the Human Factors of Scholarly Communities Supported through he Internet and the World Wide Web. JASIS, 48, 11, 987-1003.

Ickes, W. (Ed.) (1997) Empathic Accuracy, New York: The Guildford Press.

Kautz , H., Selman, B. and Shah, M. (1977) Referral Web: Combining social networks and collaborative filtering. Communications of the ACM, 40, 3, 63-65.

Konstan, J. A., Miller, B. N., Maltz, D. M., Herlocker, J. L., Gordon, L. R. and Rieddl, J. (1997) GroupLens: Applying collaborative filtering to UseNet News, Communications of the ACM, 40, 3, 77-87.

Levenson, R. W. and Ruef, A. M. (1992) Empathy: A physiological substrate. Journal of Personality and Social Psychology, 63, 234-246.

Pitkow, J. and Kehoe, C. (1997) GVU’s WWW User Surveys http://www.cc.gatech.edu/gvu/user_surveys/survey-10-1997/

Preece, J. (1998) Empathic Communities: Balancing Emotional and Factual Communication. To appear in Interacting with Computers. The Interdisciplinary Journal of Human-Computer Interaction. Elsevier.

Preece, J. and Ghozati, K. (1998) Empathy on the Internet: A Review of On-line Communities (Technical Report, Information Systems Dept., University of Maryland Baltimore County, available from the author).

Resnick, P. and Varian, H. R. (1997) Recommender Systems, Communications of the ACM, 40, 3, 56-58.

Rheingold, H. (1993) The Virtual Community, Reading, MA: Addison-Wesley Publishing.

Sproull, L. and Kiesler, S. (1991) Connections: New Ways of Working in the Networked Organization, Cambridge, MA: MIT Press.

Tannen, D. (1994) Talking from 9 to 5, New York: William Morrow and Company, Inc.

Turkle, S. (1995) Life on the Screen, New York: Simon & Schuster.

Whittaker, S. (1996) Talking to Strangers: An evaluation for the factors affecting electronic collaboration, Proceedings of ACM CSCW’96 Conference, 409-418.

Whittaker, S. and Sider, C. (1996) Email overload: exploring personal information management of email, Proceedings of ACM CHI’96 Conference, 276-283.

Whittaker, S., Isaacs, E., O’Day, V. (1997) Widening the Net. Workshop Report on the Theory and Practice of Physical and Network Communities. SIGCHI Bulletin, 29, 3, 27-30.

We gratefully acknowledge permission to use the artwork of Mark Kostabi, Kostabi World, 600 Broadway, New York, NY 10012.