AssistancePlus is a communication tool that uses co-manipulation of 3D representations of products, audio/video communication, and co-browsing to support advice-giving on pharmaceutical products over distance. Building on Clark’s theory of communication and specifically his theory of common ground [^32-34], we believe that using interactive virtual 3D objects together with audio/video communication will support communicative efficiency by creating a sense of co-presence between the patient and advisor. Our hope is that this will translate into a natural feel to the communication, as well as providing a high level of expressiveness.

AssistancePlus has been tailored to meet the specific needs of the 2 parties involved in the advice-giving interchange: the advice seeker and the advisor. The advice seeker is typically a novice and a beginner—a novice to the content domain and a beginner at using the communication tools at hand. The advisor is an expert and a professional—an expert in the content domain and a professional at using the communication tools (see [³⁵,³⁶] for a similar description of the advice seeker–advisor relationship). Due to the different nature of the 2 roles, their needs and wants are different. The advice seeker wants a service that is easy to access and easy to use. The advisor wants a service that allows him or her to communicate efficiently and effectively. This is achieved easily enough by providing them with different versions of the user interface. However, although it is true that most of the information in an advice-giving setting flows from the person giving advice to the person receiving advice, it is also necessary to have a flow in the other direction. No advice-giving session should be considered complete until the advice seeker has internalized the given information—that is, not only has received the information, but also has understood it and is motivated to act in accordance with what has been learned (see, for example, Clark [³²,³³]). It must be the responsibility of the advisor in the role of expert professional to make sure that this is so, and to do this the advisor needs feedback from the advice seeker. Therefore, it is important that advice seekers also can be expressive so that they can signal or demonstrate their understanding back to the advisor. The challenge here is to provide the right level of expressiveness for each party and, for the advice seeker, to find an appropriate balance between ease of use and expressive power. Details on how this challenge was approached in the design of AssistancePlus can be found in Multimedia Appendix 1, together with a technical description of the service.

Figure 1 shows a screenshot of AssistancePlus with the first version of the 3D player being used to demonstrate handling instructions for an asthma inhaler. The main features of AssistancePlus are (1) co-manipulation of shared 3D objects representing pharmaceutical products, (2) two-way text and audio/video communication, and (3) co-browsing of webpages with synchronized page scrolling and remote cursors

The value of 3D-supported advice-giving was assessed in 2 separate studies, both set up in the same way. Study participants (advice seekers) were exposed to AssistancePlus in a realistic usage situation after which they were questioned about their experience. The first study was explorative and probed the general usefulness of 3D visualization for advice-giving purposes. The second study was focused on the details of how 3D visualization contributed to raising the level of expressive power. A telephone session was included in the second study to provide an explicit (and controlled) reference point with which to compare their experience with AssistancePlus. The type of questions that AssistancePlus is appropriate for was also examined. A large amount of data was collected in the 2 studies, of which a selection is presented here (the full range of the data is presented in [³⁷]).

The objective of this substudy was to broadly explore the issue of how 3D visualization can be used as a tool for advice-giving on pharmaceutical products.

The objective of this substudy was to compare the experience of using AssistancePlus with that of the telephone; to identify the type of questions for which use of AssistancePlus is appropriate; and to investigate the dynamics of how 3D visualization contributes to communicative expressiveness.

For each of the 7 factors on which telephone and AssistancePlus were compared, a set of preference scores were calculated by subtracting the rating given to the telephone from that given to AssistancePlus (a positive preference score thus indicating a preference for AssistancePlus and a negative score indicating a preference for the telephone). Wilcoxon signed rank test was used to assess the statistical significance of the preference scores. Friedman tests were used to assess whether the complexity level of the product affected the choice of communication channel. Descriptive statistics were calculated for other survey data.

Being accustomed to telephone consultation, the 2 advisors could make an experience-based comparison between using the telephone and using AssistancePlus to give advice. The advisors welcomed the power of expression the 3D content provided. They felt that this extra expressiveness made it possible to handle more complex issues than could be dealt with using the telephone. In particular, the advisors valued the potential for increased precision, for example that they could rotate and zoom in on the model and then use their remote cursor to point to the exact detail they were referring to in their spoken communication.

The participants in the role of advice seekers were also pleased with their experience of AssistancePlus, both in general and with the 3D features specifically. All of the participants found the 3D features useful as a pedagogical aid and found the experience informative. The reasons they gave varied in wording, but centered on issues of clearness and distinctiveness. Several of the participants (7/10) also described the interaction as similar in feel to that of a face-to-face conversation concerning both level of expressiveness and sense of personal contact.

One somewhat surprising finding was that the activity level of the advice seekers was seemingly not very high. None of them attempted to use the 3D controls themselves and only 1 used the remote cursor purposefully. Reasons for not being more active given in the interviews were that they felt they might disrupt the advisor (6/10 participants), that the content “belonged” to the advisor (5/10), and that they felt a bit intimidated by the many buttons on the 3D player control panel (3/10). However, the main reason was that they did not feel it necessary to be more active than they were. They were active with their voice and, if they wanted to see a different view of the 3D model, they could simply ask the advisor. This said, the advice seekers wanted to keep the feature (9/10). It seems that knowing that they had the option to control the model made them feel more involved in the advice-giving process (even if they did not actually do this). Similarly, the advisors felt that, while the vocal feedback the advice seekers gave was sufficient for the advisors to judge understanding in the present situation, in other situations it might be helpful having the advice seekers control the model themselves to demonstrate what they have learned.

Figure 4 shows the median value, interquartile range, and total range of the preference scores that were calculated by subtracting telephone ratings from AssistancePlus ratings for each of the 7 factors on which telephone and AssistancePlus were compared (see Methods section). All the preference scores were positive in AssistancePlus’ favor, and all scores except ease of use were shown to be statistically significant (Table 1). For ease of communication, personal contact, and trust, the magnitude was moderate; for explanatory power, understanding, and efficiency it was more pronounced. The results were corroborated in the interviews, where participants described their experience with AssistancePlus as being more clear and expressive, being more personal and present, and giving a deeper sense of understanding. Also, several participants described the feel of the interaction when using AssistancePlus as being more similar to a face-to-face meeting than that of a telephone conversation. It should be noted, though, that both the questionnaire and the interview were administered in conjunction with the AssistancePlus session, which might have introduced a positive bias in AssistancePlus’ favor, being the more recently experienced.

Figure 5 displays the results of the preference ratings given for the 4 types of products and 4 different channels. On type of product (front to back in Figure 5), the variation was found to be significant for all product categories except complex medications (Table 2). On type of channel (left to right in Figure 5), AssistancePlus and Internet differed significantly, but telephone and pharmacy store did not. The results indicate that the advice seekers prefer a simple channel for simpler questions and a more powerful channel for more complex questions—that is, that they make their choice with consideration to the complexity of the issue at hand.

Somewhat surprisingly, though, the advice seekers did not rate the face-to-face setting—ostensibly the most expressive channel among the alternatives—as the preferred channel for the most complex type. In fact, the results instead showed an inverse relationship between the pharmacy store and product complexity. The reason for this was revealed in the interviews. The participants saw the meeting in the pharmacy store not as an expressive and personal face-to-face interaction, but as a fleeting encounter in a nonprivate, noisy, and stressful environment. The telephone received surprisingly low ratings for all categories indicating that the telephone is simply not a popular alternative when seeking medical advice.

The level of activity on the part of the advice seekers was very low in the first study, with none of them even attempting to use the 3D controls. This behavior changed in the second study, where they were more active. Typically, the actions they performed were simple ones such as using their remote cursor to point to details on the 3D model (16/23, 70% of participants), moving the playhead to go to a specific point in the animation (2/23, 9%), and starting and stopping playback (3/23, 13%). Only 1 participant used the click-and-drag controls to rotate and zoom in on the model. However, the preference for simpler action was not because they did not know how to perform these more advanced actions. As in the first study, they simply found it easier to have the advisor do it for them. Still, the participants were very positive about having the option to be able to control the 3D content themselves (N = 23, mean [SD] 5.42 [0.90], range 1–6) and wanted to keep the feature. The advisors were also pleased that the advice seekers had the option to control the 3D content themselves.

Advice seekers and advisors alike found AssistancePlus useful as a pedagogical aid for advice-giving and particularly appreciated the clarity and exactness brought to instructions by the 3D features. This does not mean that AssistancePlus is the right choice for all types of advice-giving situations. The results from the second study indicate that information seekers do not just go for the most expressive channel, but differentiate their choice. Perceived level of expressiveness is one factor affecting choice, but other factors are also important. For instance, the pharmacy store was an unpopular alternative for the reason that it was seen as lacking in privacy. If we try to isolate the issue of level of expressiveness here, it seems that participants preferred the channel that offers not too much, not too little, but just the right level of expressiveness. The implication is that we should provide advice seekers not with a single catch-all channel, but with a selection so that they can choose the right channel for the question at hand. Of course, the downside of providing choice is that it places a burden on the advice seeker to choose among an ever-increasing range of channels. We believe that a major future challenge will be to provide guidance services to help advice and information seekers to find not only the right information, but also the right channel.

In the advice-giving situation, most of the communication naturally flows from the person giving the advice to the person receiving the advice, but we have argued that it is important that the advice seeker provide feedback so that the advisor can properly judge understanding and motivation. Here, care must be taken not to overwhelm or intimidate the advice seeker, who perhaps for the first time is communicating over distance using advanced technology. Our solution is to provide self-use of, for example, the 3D controls as an option, something that one can do, but does not have to do.

An interesting—and we believe important—finding in this area was that advice seekers did feel more involved when presented with the option of using the 3D controls, but this sense of involvement was not the result of actual action. Instead it seemed to stem from the awareness that they could be active. Both those who actively used the 3D controls and those who did not felt empowered by knowing they could control the 3D content (theoretical support for this empowerment effect can be found in Clark’s theory of common ground [^32-34]). This finding suggests that it is positive to present advice seekers with more functionality than they are expected to use. However, we hypothesize that the empowerment effect will occur only if the functionality is perceived as something they can manage. This means that either the functionality needs to be simple enough to be self-explanatory, or it can be easily explained by way of demonstration, which is the case with the playback and 3D controls used in the studies. We believe that this finding, provided that it can be corroborated in further studies, can have important consequences for the design of online communication services of the kind presented here.

In this paper we have exemplified how 3D visualization can be used in health communication for a specific advice-giving situation. It is in the custom tailoring to the specific communication setting and to its actors that the service gains its edge, but we believe that the concept it presents—ease of access with low technical and usability thresholds and a level of expressiveness tailored to the demands of the specific communication situation—is relevant for any area of health care where there is a need for improving access to information and for bringing patients and caregivers closer to each other.

It must be noted that both presented studies used convenience samples and were performed in usability laboratories. There may also be a novelty effect, and it remains to be seen how a service such as AssistancePlus would work in a real-world setting with real patients and caregivers who might react differently from the test participants in our studies. That said, the results are promising and we believe AssistancePlus can provide both an alternative to the telephone when it is not expressive enough and a complement to the face-to-face interaction when a physical meeting is not convenient.

We have not performed a detailed analysis of the cost of setting up and running a service such as AssistancePlus, but it is our assessment that it should be comparable with that of setting up a telephone-based call centre. Of course, the easiest way to get a service like AssistancePlus up and running would be to add it to the list of services offered by an existing call centre. Since all the functionality demonstrated in the AssistancePlus application can be implemented using open source free software, this means development and running costs can be kept down and, while the production of the 3D content is an extra cost for each product that is added, it is a cost of moderate size.

We have shown that 3D visualization techniques paired with video conferencing can be useful in supporting advice-giving over distance and can be used even in low-bandwidth settings. We also showed that 3D visualization brings clarity and exactness to given instructions and contributes to communicative efficiency. It is a technology that can be made available to the general public and, if implemented with proper consideration to the role and character of the advice seeker, will be both easy to access and easy to use and, in some cases, even preferred to face-to-face communication.