Designing and Implementing Virtual Patient Support Communities: A German Case Study [1]

Virtual communities can – theoretically – be described as a solution for meeting ubiquitous information and interaction needs. Such needs occur in healthcare, when for example freshly diagnosed cancer patients develop very strong information and interaction needs. But how can such a platform for cancer patients be designed, implemented and introduced practically? What specifications must a possible technical infrastructure meet? The COSMOS project addresses these topics. Besides these aspects, new technological possibilities like mobile services and mobile devices are influencing virtual communities. Subjects like ubiquitous community access, new possibilities of user identification, and location-related services are of special interest since they might allow real anytime-anyplace access to the community platform. Thus new, value-adding services to community members could be added. But nevertheless: Only socially accepted, technically stable and economically feasible solutions can ensure sustainable success of (mobile) virtual healthcare communities. Virtual Patient Support Communities, p-2 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. Introduction Virtual Communities (VC) can be defined as a group of people who gather because of a common interest, problem or task and whose members are independent of time and space for interacting (for similar definitions see e.g. (Armstrong & Hagel, 1996; Mynatt, Adler, Ito, & O'Day, 1997; Preece, 2000; Schubert, 1999). Therefore virtual communities have great potential to serve ubiquitous needs. Such an omnipresent problem situation exists for instance in healthcare, when patients develop a desire for information and communication exceeding the resources of the treating physician. From the characterisation of virtual healthcare communities as ubiquitous information and interaction spaces, we have derived design requirements for VCs as well as for the process of developing a community platform. The results are transferred into the healthcare domain, particularly into the situation of cancer patients. On this basis we give an overview of our experiences with the development of a community platform for cancer patients. In closing we put special emphasis on the possibilities and challenges of mobile technologies for virtual healthcare communities. Research Design and Research Plan Research design describes the key objectives of the research project, what methods will be used for data collection and analysis as well as how the research process shall take place. The objective of this venture is to plan, build, introduce and evaluate IS-platforms for cancer patients (see Figure 1). --Figure 1 Goes About Here --As usual in pilot projects, the starting-point of this research is a socio-organisational problem (in this case the situation of cancer patients after they leave hospital). At the beginning, an analysis of cancer patient’s situations was performed in-depth through literature review and above all case studies, using interviews, questionnaires, observations and document analyses (Yin, 1989). The results of the analysis were used for designing an IS-platform. This platform was implemented in the field, and improvements in the system are made during the remainder of the project. At all times, on all levels, a continuous evaluation takes place and thus iterative learning by the system developer can be augmented at all stages. The objective of the field studies was to study cancer patients’ needs and to analyse already available web-based offers for cancer patients. Therefore we analysed the situation of Virtual Patient Support Communities, p-3 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. cancer patients in general with a standardised questionnaire, followed by in-depth studies in 5 different cancer self help groups with approximately 100 active members. Methods applied here were ethnographic analysis, semi-structured interviews, observations and document analysis. After that, web-based information and interaction opportunities were investigated. Finally, identified cancer patients’ needs and already available interaction opportunities were compared. On this basis the development of the prototype began, using an iterative process model in order to allow a high degree of user involvement in the development process. After several iterations the prototype was introduced to public on August 18 2002. The platform is being tested continuously by users and additionally by focus groups in controlled settings. Field Studies Needs Analysis of Cancer Patients Patients’ needs and demands for information often increase after a diagnosis of a disease or during medical treatment (Sheppherd, Charmock, & Gann, 1999). These demands can be the result of asymmetric information, for instance between patients and physicians. Patients may seek information to help them make sense of a cancer diagnosis or to provide them with information which assists them in making informed decisions about treatment. Besides demands for factual information, there can be a desire to seek emotional support and to communicate with other patients. These behaviours, such as participating in self-help groups, can play an important role in dealing emotionally with a disease (for an overview see Hasebrook, 1993). If we assume that there is a correlation between the threat to quality of life imposed by a cancer diagnosis and the need to seek and obtain knowledge and support (e.g., Eysenbach, 2000), the potential benefit of cancer-related communities becomes evident. However, the diversity of over 100 types of cancer, the diversity of the presentation of the same disease in two patients, the complexity of treatment modalities coupled with the hardly manageable extensive professional and lay literature in this area, all combine to make coping with cancer, even through participation in virtual communities, extremely difficult. Recent research on cancer patients’ informational demands (e.g., Kaminski, Thomas, Charnley, & Mackay, 2001) demonstrates a strong information interest in the following areas: Side effects/how I will feel; Explanation of disease and prognosis; Treatment options and Virtual Patient Support Communities, p-4 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. explanations of therapy; Logistical issues (transportation, work, etc.); Lifestyle issues (exercise, diet, sexuality, smoking); Follow up/what happens after therapy finishes; and Support or self help groups, alternative medicine. Above these information demands, books of cancer survivors such as Lance Armstrong (2001) and their huge success show that cancer patients also have strong desires for emotional support and empathy. Many of patients’ needs to find answers to perceived and real problems as well as informational needs can be solved through self help groups. Self help groups exist in many major cities and for different types of cancer. According to information provided by the AOK, Germany’s largest health insurance company, only approximately 5-10% of cancer patients take part in self help groups. Reasons for this can be, among others, that interested patients are unable to locate a group in their vicinity or that meeting times of groups do not fit the individual schedules. Very often integration into a group plays an important role, and, in particular, fear and mistrust in “strangers” are often experienced. Taking part in self-help groups is linked with talking about a very intimate subject like one’s disease and presents, and therefore is a very uncomfortable situation for most patients. These circumstances of real-life communities on one hand and the demand for information and interaction on the other show the great potential of virtual healthcare communities as ubiquitous information and interaction spaces for solving these problems. Broadbased and enlightening information showing advantages and disadvantages of alternative treatments as well as infrastructure issues such as, where one can obtain what type of treatment is not provided. In order to enable patients to be autonomous in their decision–making, a solid basis in prior related research is necessary. Empirical Findings of Conducted Field Studies During the European Week against Cancer (October 2002) a standardised questionnaire was distributed to approximately 500 visitors, with 116 responding. More than 60% of the cancer patients agree with the question that their physician has a big information advantage. One third of them agree in parts to the statement that they want to look for information on the Internet. 28.8% of the female cancer patients use mobile phone as well as the Internet, by men mobiles phones are used by 56.3% and the Internet by 43.8%. More than two third agreed in parts that Virtual Patient Support Communities, p-5 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. they want to communicate more with other patients. These statements emphasise the need for web-based or mobile information and interaction services. Additionally we conducted ethnographic analysis of self-help groups. We conducted narrative interviews with self-help group leaders and several semi-structured interviews with the group members. The results showed that there is a lack of information and interaction possibilities for cancer patients and their relatives. The patients want and need information about their specific type of cancer, treatment or hospital. Most of the members of self help groups found it important to share information and to speak with other patients in the same or similar situations. They got hope and encouragement when they saw one of the members recovering. Even if the meetings just took place quarterly, the members called each other by telephone very often to get advice from others. Some members travelled a long way to their self help group meetings. The use of materials and tools like mobile phones, personal digital assistants or the internet correlated with the average age of the group and the age of the members. In some groups almost all members had mobile phones and used the Internet. Unrestricted access to valid, understandable and relevant information as well as the possibility to contact other patients of the self-help group at any time was considered as very important. But this approach is only possible for members of self help groups and therefore not for the majority of cancer patients. Thus our results confirm cancer patients’ demands for information and interaction services. Analysis of Web-based Information and Interaction Offers for Cancer Patients Until the start of the project (mid 2001) a working virtual community for cancer patients could not be identified on a European level (German or English speaking). Only in the US were we able to find first approaches towards virtual cancer patient communities (Daum, Klein, Leimeister, & Krcmar, 2001). Existing internet services in German language offered mostly content without user interaction possibilities. Additionally hardly any services focused on quality assurance of their content or target-group specific editing of their content (e.g., most offered medical content was almost not understandable by laypeople). Concerning mobile services support for access through mobile devices, we could not identify any services in the German speaking Internet. Therefore a pilot project for developing a community platform for cancer patients was initiated. Virtual Patient Support Communities, p-6 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. Requirements and the Development Process The Dilemma of Requirements Collection Although software development is usually done within an organizational framework, many of the developed information systems do not match the needs of the target users. Many orphaned community platforms on the internet underline this. Traditionally system development takes place in a linear manner, starting with the conception phase and ending with the phase of the death or substitution of the system. Many alternative models have been developed since the linear model holds many dysfunctional aspects (see Boehm, 1988). The internet as a possible environment for an information system, especially with its heterogeneous user groups, demands more flexibility and has a high degree of uncertainty concerning how hardware and software will be used. The collection of requirements often takes place very early in the development process and is transferred into a requirements specification. Little attention is paid to the alteration or adaptation of already acquired requirements during the following phases. There is little time and space to pick up on new requirements in most processes. Mistakes within the requirements specification and the management of customer requirements reduce system quality and generate high costs for ongoing system corrections. Therefore the collection of requirements should be emphasized and the target user should be involved in early stages of the software development cycle. The requirements engineering approach tries to fulfil this request. Requirements engineering is defined by the IEEE Std. 610.12 as “(1) the process of studying user needs to arrive at a definition of [...] requirements; and (2) the process of studying and refining [...] requirements” (1990, cited in Hoffmann, 2000). It also emphasizes the fact that requirements can change during the development process. In order to develop a platform that meets users’ needs, requirements engineering addresses many essential success factors for system development and is therefore a suitable framework for the development of a community platform in general. There are many activities and methods reported in the literature that guide and support the discovery of requirements (for an overview see Hoffmann, 2000), but are they applicable for our purpose and can they be combined with a process model for system development? Virtual Patient Support Communities, p-7 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. Although considering many published approaches [2], we did not find an appropriate model that combined both an applicable process for the development of a platform on the internet and the inclusion of requirements engineering activities that involve intense participation of targeted users, and certainly none with a proven track record. Requirements Deducted from Prior Field Studies The results of prior field studies about the information and interaction needs of cancer patients (see above or Leimeister, Daum, & Krcmar, 2002) and the examination of existing web based information platforms on the German language market (see above and Daum et al., 2001) lead to the general requirements listed in Table 1. --Table 1 Goes About Here --Starting with these first requirements, the process model for community platform engineering has to allow the integration of parallel activities of requirements engineering in order to provide the flexibility needed for the development of a platform for cancer patients on the Internet. Therefore an applicable process model should: be an iterative process, be able to adapt to changes of requirements during the development process, include several builds of prototypes, be easy to apply especially for small and medium size projects, be applicable for different types of services (information as well as interaction), and involve users and/or experts from the beginning. The following suggests a process model that meets these preconditions. A Community Platform Engineering Process (COPEP) The objective of this section is to describe a process model for developing an information system, a community platform for cancer patients. Since system requirements are neither completely nor exactly defined, a linear model does not fit the uncertainty that arises from the field. An iterative model seems to be more appropriate. Starting out with general requirements, the system can be built step by step. The outcome of each stage of the iterative development should be evaluated. Within each iteration, however, the type of development should be shaped by the demands of the situation. Figure 2 shows the process model COPEP (Community Platform Engineering Process) that was used during the development of the Internet platform www.krebsgemeinschaft.de for the target group of breast cancer patients. Virtual Patient Support Communities, p-8 1. Leimeister, J. M.; Krcmar, H. (2006): Designing and Implementing Virtual Patient Support Communities: A German Case Study In: The Internet and health care: Theory, research and practice. Hrsg. / Editors: Murero, M.; Rice, R.E. Publisher / Verlag: Lawrence Erlbaum Associates, Erscheinungsort/ Published in: Mahwah. Erscheinungsjahr / Year: 2006. --Figure 2 Goes About Here --The heart of this model is an iterative process, adapted from the generic spiral process model (Boehm, 1988; Wigand, Picot, & Reichwald, 1998). It is combined with a prototyping approach. Each iteration consists of four phases: planning, analysis, engineering and evaluation. Different from the original spiral model, a much stronger focus has to be put on the building of prototypes and the involvement of users in evaluation. The goal of the engineering phase of each iteration is the building of a prototype in order to get a tangible version of parts or the whole product very early in the development process. After the evaluation phase, the second iteration starts over with planning again, but applying experiences from the previous iteration.