Training in outbreak investigation through use of an online discussion group.

Introduction

Training and retraining of the estimated 450,000 U.S. public health employees (Bureau of Health Professions, 2001) at the local, state, and federal level are challenging tasks. The wide variability in student background (i.e., education and professional experiences), the recurring need for training because of staff turnover, and the lack of time and resources increase the difficulty of providing training that is relevant, practical, and cost-effective.

Distance learning, in which the instructor and student are separated by time, location, or both, has been promoted as an effective tool for training the public health workforce. Involving approaches such as print-based materials, computer- or Web-based courses, or videotapes, distance learning can be designed so that students can fit the instruction into their schedules at their convenience. It can be designed to allow students to progress at their own rates, skip over topics they already know (or do not need to know), or repeat topics with which they have difficulty. In addition, distance learning can provide the same information and training simultaneously to geographically dispersed audiences and repetitively over time so that a reasonable level of consistency can be achieved.

Distance learning also has its shortcomings, however. Some distance learning can seem remote and impersonal to the student, particularly when limited or no human contact is provided by an instructor or other students (Moore, 1989). Opportunities for the student to apply the training material to his or her specific work situation can be lacking, and this circumstance may hinder the student's assimilation of the material. Furthermore, some formats of distance learning lack spontaneity and the ability to redirect training according to student needs as the training evolves. For example, in a simultaneous nationwide satellite broadcast, the instructor has limited capabilities for changing the flow of the training during the event, and students, for the most part, have no way to indicate that a change in the flow is needed.

The "Botulism in Argentina" computer-based case study was developed by the Centers for Disease Control and Prevention (CDC) as a distance-learning approach to teach public health workers knowledge and skills needed to conduct an outbreak investigation. "Botulism in Argentina" is an interactive, self-instructional course in which students work through a real-life outbreak investigation. Initial evaluations of "Botulism in Argentina" suggest that it is effective in teaching knowledge and skills needed in outbreak investigation and could be a model for other training efforts (Stehr-Green & Gathany, 2002). As a self-instructional course, however, the case study lacks benefits that result from student-instructor and student-student interactions.

CDC conducted a pilot in which the "Botulism in Argentina" computer-based case study was supplemented by an online discussion group to address the lack of human interactivity in the self-instructional format. The goal of the pilot was to determine the feasibility of conducting online discussions with public health employees; considerations included ease of use, student/instructor satisfaction, and resource requirements.

Methods

Participants for the discussion group were selected from members of the target audience for the "Botulism in Argentina" computer-based case study (i.e., public health practitioners with knowledge of basic epidemiologic and public health concepts who wanted to learn useful skills in investigating outbreaks). Participants were recruited through the membership of the National Environmental Health Association (NEHA) by word of mouth and through an ad posted on NEHA's Web site. Background information (e.g., demographics, education, and work experience) was collected from all participants before the online discussion was initiated.

Two instructors (i.e., an epidemiologist and an instructional designer) facilitated the discussion, which was divided into six steps (to correspond to the steps of the case study). They developed a syllabus outlining the cutoff dates for discussions for each step of the case study, posted thought-provoking questions and assignments for each step, summarized participant postings on selected topics, and tried to maintain participant interest through intermittent e-mail reminders. Two expert consultants were available to address questions raised by participants regarding the practice of environmental health.

The online discussions were conducted through the Case Study Forum established for this project with WebBoard (Version 4.2) Web discussion forum and conferencing software located on a CDC server. The WebBoard allows participants to communicate with each other on the Internet in an asynchronous manner by posting e-mail-like messages. Participants can access the online discussion anytime day or night and can do so from any computer with Internet access.

Shortly before initiation of online discussions, participants downloaded the "Botulism in Argentina" computer-based case study and installed it on their computers. Participants received written instructions on how to access and submit comments/questions to the online discussion and the discussion syllabus. To officially kick off discussions, an audio conference was held so that participants could introduce themselves and ask questions about the pilot test.

Outcome Measures

The authors measured participation rates by counting postings to the online discussion and summarized them by participant and for the entire group. To determine the timeliness of the postings, the date of each posting was also compared with the cutoff date for the step noted in the syllabus. Level of participation for participants with different characteristics (e.g., age, gender) was compared. Participant satisfaction was evaluated after the online discussions were completed. Participants were asked to complete a survey in which they rated their level of agreement (on a scale of 1 [strongly disagree] to 5 [strongly agree]) with a set of positive statements about the training experience. They were also asked to describe specific problems they had had with the online discussion.

[FIGURE 1 OMITTED]

Results

Description of Participants

Eighteen target audience members from across the United States agreed to participate in the pilot test (Figure 1). Eleven participants (61 percent) were male; the median age was 36 years (range: 22-53 years). Most participants (66 percent) were environmental health specialists, sanitarians, or food safety inspectors. Participants had a median of 4.5 years of experience in public health (range: 1-33 years). All but three participants considered themselves to have "average" computer skills. Two considered themselves beginners and one an expert. All participants had used e-mail, 10 (56 percent) had participated in a listserv, and five (28 percent) had participated in a Web-based discussion group.

Level of Participation

The audioconference was held on April 23, 2003, and was joined by 11 participants and both instructors. The online discussion occurred from April 28 through June 8, 2003. Of the 18 participants, 16 (89 percent) posted at least one message to the online discussion. Of the two who posted no messages, one reported that she had been misinformed about the course content and had decided not to participate. The other could not be reached.

Overall, 166 messages were posted during the six weeks of the pilot test; 113 (68 percent) of the messages were posted by participants. Most of the participant postings were responses to questions or assignments made by the instructors; about one-quarter of their postings specifically referred to a posting by another participant (e.g., responded to questions from another participant, added to or supported the thoughts of others) or initiated a new thought or discussion thread. Content of the postings frequently referred to the participant's own professional or personal experiences.

The median number of messages posted per participant was 5.5 (range: 0-16). The number of messages posted per participant did not differ by gender, age, years of public health experience, or previous participation in a listserv or discussion group. The number of postings was higher for participants who had previously participated in investigating an outbreak (median: 7.5 postings) than for those who had not participated in an outbreak investigation (median: 4.0 postings) (p = .08).

[FIGURE 2 OMITTED]

The number of messages posted decreased as the pilot progressed. The highest number of messages (58) was posted for Step 1 of the case study. The lowest number was for Step 5 (12 messages) (Figure 2). The posting of messages for each step of the case study followed the discussion syllabus only loosely. Over one-third of the postings for particular steps occurred after the cutoff date as noted in the syllabus.

Evaluation of Experience

Seventeen (94 percent) participants submitted an evaluation of the online discussions. On a set of standardized statements, the participants were positive about the learning experience (Table 1). In general, participants believed that the online discussion added value to their learning experience, which otherwise would have been self-instruction only. All participants enjoyed reading other participants' postings and thought that other participants brought up questions and issues they had not considered. A number of participants said that they were initially uncomfortable posting their ideas online but that these feelings changed as the discussions progressed. Several participants described technical difficulties accessing the online discussions and posting messages. One participant reported difficulty viewing the list of postings and seeing which ones had already been read. One participant (and one consultant) reported that messages they posted had been lost (probably because they did not "post" them before signing off the system).

Discussion

Although the authors did not evaluate whether individuals who participated in the online discussion of the "Botulism in Argentina" computer-based case study learned more than do individuals who complete the case study on their own, the findings suggest that the combination of the case study and online discussion was a positive experience overall. First, there was a high level of participation by most members of the discussion group; participants appeared to be reading the postings of their colleagues and independently posed questions to the group or sought guidance from other participants. Second, the vast majority of postings were thoughtful and relevant to the discussions, often referring to the participant's personal experiences with regard to the subject under discussion. Third, almost every aspect of the learning experience was rated highly by participants in the satisfaction survey, and all but three participants (18 percent) said that they would participate in a similar undertaking in the future. Finally, the discussions clearly supplied the human interactivity (student-instructor and student-student) missing from the self-instructional case study. As evidenced by selected statements quoted below, the participants enjoyed these interactions:

* "I had no idea that I would have access to all this input from other sanitarians across the country. The mixture of less experienced and very experienced participants was very helpful, also the mix of sanitarians and epidemiologists. I wish the same group could do more exercises together."

* "The discussion met my expectations. I expected to see a different answer from every person for every question. This definitely happened!"

* "I definitely did not feel alone, more like a team going through the investigation."

* "I enjoyed reading participants' responses [and] the gentle Socratic 'push' given by the instructors."

* "The role of the moderators can't be overstated, and the job done by [the moderators] was superb."

The online discussion seemed more successful than those reported in the literature for health care providers, a group seemingly similar to these pilot participants (Chan, Leclair, & Kaczorowski, 1999; Sargeant et al., 2000). Why was this? The authors believe it was largely due to the characteristics of the participants and the self-selection involved in the identification of the discussion group. Research suggests that students who are voluntary students, motivated with higher expectations, older, and more serious in their studies tend to be more successful in online discussions (Palloff & Pratt, 1999). These characteristics easily fit the discussion group of young and middle-aged adults who volunteered for this pilot because they wanted to learn skills that were relevant to their job performance.

The computer-based case study (as the focus of the discussions) probably also contributed to the success of the undertaking. The case study, which has an engaging interface and the feel of a simulation, seems to draw the student in (Gathany & Stehr-Green, 2003). Students seem motivated to complete the case study and are interested in talking with others about the concepts raised.

The nature of the discussions, no doubt, also played a part in the success. Palloff and Pratt posed six elements as critical to the success of online learning: 1) a sense of safety and trust, 2) responsiveness from instructors and other participants, 3) relevance and ability to relate the subject matter to professional and life experiences, 4) respect between participants as people, 5) openness and an ability to freely share thoughts, and 6) empowerment to impact the training experience (Palloff & Pratt, 1999). The online discussion in this study fulfilled many of these elements. Participants were courteous, supportive, and respectful of each other. The instructors made a concerted effort to respond to postings by individuals, and clearly the participants were reading each other's postings. The online discussion provided many opportunities for the student to apply the concepts addressed in the computer-based case study to his or her specific work situation, and participants frequently referred to their own work experiences and made generalizations from those experiences. The online discussion also was flexible, so the training provided in the computer-based case study could be supplemented to meet the current needs of the participants.

There were problems with the online discussions, however:

* The Web conferencing software (WebBoard, Version 4.2) used for the online discussions was cumbersome. Although somewhat intuitive for individuals who had participated in online discussions before, the software had idiosyncrasies that were difficult to overcome. For example, participants could not view the titles of all of the posted messages on entering the system, but needed to expand selected "discussions" individually. Therefore, participants had to work to see which messages were new or of interest. In addition, the marking of messages as read (somewhat dependent on the account settings of the individual participant) was not specific to individual messages that had actually been read. Instead messages were automatically marked as read if the participant logged off, whether or not they had been read. Some participants reported difficulty accessing the discussions and posting messages, and some even lost messages they had composed. It is likely that all of these problems led to participant frustration, decreased comprehension of the information, and caused the loss of contributions to the discussions. Therefore, the authors agree with other online-discussion organizers that software that can be easily used by participants who lack computer proficiency or comfort with computers is needed to enhance the experience for all participants (Palloff & Pratt, 1999; Sargeant, et al., 2000).

* A few participants (and the instructors, at times) were overwhelmed by the discussions and the extensive comments made by others. Given the lengthy period over which discussions occurred, the number of new postings each week, and the fact that participants moved through the case study at different paces, there was the potential for participants to lose the thread of the ongoing discussions. This problem would be heightened for any participants unable to visit the site regularly and review messages more frequently. Shortening the discussion period, adhering to the syllabus, and decreasing the number of participants might lessen this problem, as participants would move more in concert. These changes, however, could negatively affect the quality of the discussions as well as increase the perstudent cost for the instruction.

* Maintenance of adequate levels of participation was a looming concern. The instructors tried to promote participation by sending e-mail reminders, making specific assignments, and posting controversial messages meant to evoke responses; nonetheless, postings fell over time. Although this result is understandable given the busy work and personal schedules of participants, it is unfortunate. Award of a certificate or continuing-education credits tied to a requirement for a certain level of participation might increase participation, but it also might negatively affect the nature of the discussions and the general "feel" of the group interactions.

Resources used to conduct the online discussions can only be estimated. They include conferencing software, Web support, server access, audio call costs, and instructor and participant time. The instructors spent well over 15 hours enrolling participants, setting up participant accounts to access the online discussions, and developing the course structure. They spent an additional 35 to 40 hours reviewing postings to the online discussion, responding to messages when appropriate, summarizing postings from selected topics, and following up on individual participants who had expressed a need for help on a particular topic. The participant time commitment is unknown, but, based on accounts from two active participants, may have reached as high as 20-25 hours (including three to four hours to complete the case study itself). Therefore, the authors estimate that the time required for this course was similar to that necessary for a one-week on-site course. If an on-site course had been undertaken, however, travel time and costs would have been incurred. Furthermore, the online format allowed participants to continue their normal work and provided flexibility in the timing of their participation. As one participant put it: "With my schedule, it was great to be able to move in and out of the course at any time, for any length of time."

Conclusion

In summary, the online discussion of the "Botulism in Argentina" computer-based case study was a positive experience for participants and instructors. It appears to be an enjoyable and economical way to receive training and to make contact with other professional colleagues. The authors encourage others to consider this approach for their public health training needs. They suggest, however, that close attention be paid to the problems identified in this pilot as well as to those that may be unique to the students being trained.

Acknowledgments: The authors would like to thank Stuart Capper for his advice about online discussions, Tom Dickey for recruiting the participants, Robert Harrington and Vickie Church for acting as environmental health consultants, and Wally Wilhoite and James Brooks for their technical support with WebBoard. Thanks are also due to the participants in the online discussions for making the pilot test a fun and rewarding experience.

Corresponding Author: Jeanette Stehr-Green, Consulting Medical Epidemiologist, Public Health Foundation, 6305 Blue Mountain Rd., Port Angeles, WA 98362. E-mail: [e-mail].

REFERENCES

Bureau of Health Professions, Health Resources and Services Administration. (2001). The public health work force: Enumeration 2000. http://www.phppo.cdc.gov/owpp/docs/library/2000/Public percent 20Health percent20Workforce percent20Enumeration percent202000.pdf (5 Oct. 2004).

Chan, D.H., Leclair, K., & Kaczorowski, J. (1999). Problem-based small group learning via the Internet among community family physicians: A randomized controlled trial. MD Computing, 16(3), 54-58.

Gathany, N., & Stehr-Green, J. (2003). Scenario-based E-learning model: A CDC case study. Learning Circuits. http://www.learningcircuits.org/2003/apr2003/gathany.htm (5 Oct. 2004).

Moore, M.G. (1989). Editorial: Three types of interaction. The American Journal of Distance Education. http://www.ajde.com/Contents/vol3_2.htm#editorial (5 Oct. 2004)

Palloff, R.M., & Pratt, K. (1999). Building learning communities in cyberspace: Effective strategies for the online classroom. San Francisco: Jossey-Bass.

Sargeant, J.M., Purdy, R.A., Allen, M.J., Nadkarni, S., Watton, L., & O'Brien P. (2000). Evaluation of a CME problem-based learning internet discussion. Academic Medicine, 75(10), S50-2.

Stehr-Green, J., & Gathany, N. (2002). Impact of a computer-based case study on outbreak investigation skills. Academic Exchange Quarterly, 6(3), 42-48.

Jeanette Stehr-Green, M.D.

Nancy Gathany, M.Ed.

TABLE 1 Responses of Participants to Positive Statements about the
Online Discussion

                                                        Percentage
                                              Mean      Selecting a
Positive Statement                            Rating**  Rating of 4 or 5

 1. The audioconference on April 23 was a     4.6        91%
    good way to initiate the online
    discussion pilot. (N = 11)*
 2. I enjoyed hearing about the backgrounds   4.7       100%
    of other participants during the
    audioconference. (N = 11)
 3. The document "Instructions for Online     4.4        94%
    Discussion" adequately explained how to
    participate in the online discussion.
    (N = 16)
 4. I did not have any technical difficulty   4.1        76%
    accessing the online discussion. (N =
    17)

 5. I did not have any technical difficulty   4.1        76%
    posting messages to the online
    discussion. (N = 17)
 6. I felt comfortable posting my ideas to    4.1        81%
    the online discussion group. (N = 16)
 7. When I posted a question to the online    4.0        75%
    discussion group. I always received a
    useful response. (N = 12)
 8. The online discussions highlighted        4.6       100%
    important aspects of outbreak
    investigation. (N = 16)
 9. The online discussions were applicable    4.1        87%
    to my work. (N = 15)
10. The online discussions added value to     4.6        94%
    doing the computer-based case study.
    (N = 16)
11. The online discussions were a good use    4.3        88%
    of my time. (N = 16)
12. Other participants brought up questions   4.6       100%
    or comments in the online discussions
    that I had not considered. (N = 16)
13. I enjoyed seeing other participants'      4.6       100%
    comments. (N = 16)
14. I had fun participating in the online     4.5        94%
    discussion group. (N = 16)
15. I would participate in another online     4.5        88%
    discussion group similar to this one.
    (N = 16)
16. I would recommend online discussion       4.6        94%
    groups as part of training efforts to my
    colleagues. (N = 16)

*N = number of participants responding to statement.
**I = strongly disagree, and 5 = strongly agree.