Virtual Humans Playing the Role of Patients in Self-medication

Consultations: Perspectives of Undergraduate Pharmacy Students

Ana Paula Cláudio

, Maria Beatriz Carmo

, Mara Pereira Guerreiro

3,4

, Afonso Cavaco

Vítor Pinto

and Ana Pinha

BioISI- Biosystems & Integrative Sciences Institute, Faculty of Sciences, University of Lisboa, Lisboa, Portugal

Faculty of Pharmacy, University of Lisboa, Lisboa, Portugal

CiiEM -

Instituto Superior de Ciências da Saúde Egas Moniz, Monte da Caparica, Portugal

Escola Superior de Enfermagem de Lisboa, Lisboa, Portugal

Keywords: Virtual Humans, Consultation Skills, Communication Skills, Pharmacy Students, Self-medication.

Abstract: This paper describes an interactive application conceived to improve non-prescription medicines consultation

skills in undergraduate Pharmacy students and a user test carried out with two sets of participants. Resorting

to virtual humans that play the role of patients and communicate with the students by speech and by facial

and body language, the application stimulates students’ engagement in true-to-life situations in a controllable

environment. Two usage modes are available: the training mode, used autonomously by the student, and the

assessment mode which is used by the teacher to evaluate the student’s performance. A BackOffice Web

application was also implemented to assist teachers’ work. It supports the collection of data about students’

performance, the creation of new self-medication situations and their posterior insertion in the application.

The overall opinion of the participants in the user-study was quite positive about the usefulness of the

application as a tool to improve students’ non-prescription medicines consultation skills. Moreover, some

valuable suggestions were gathered during this testing process.

1 INTRODUCTION

Self-medication is common in community pharmacy.

For example, research indicates that about a quarter

of the clients in urban Portuguese pharmacies

purchased medicines without physicians’ advice

(Martins et al., 2002). It is acknowledged that self-

medication has a number of potential advantages for

stakeholders - consumers, health professionals,

pharmaceutical industry and the health system - but it

is not without risks, namely adverse effects (Asseray

et al., 2013) (Schmiedl et al., 2014) and misuse

(Cooper, 2013). International and national

organizations, such as the World Health

Organisation, have advocated a role for community

pharmacists in assisting consumers who present

minor ailments or request non-prescription

medicines.

In his seminal work on quality of care Donabedian

considered both technical and interpersonal elements

in the performance of practitioners (Donabedian,

1988). The latter was regarded as “the vehicle by

which technical care is implemented and on which its

success depends”. In other words, a successful

consultation relies not only in clinical knowledge and

judgment but also on communication; self-

medication consultations are no exception. Therefore

pharmacy educators are faced with the challenge of

developing both clinical and communication skills in

students.

This paper describes an interactive application to

support skills training and assessment of

undergraduate Pharmaceutical Sciences students and

continues previous works from the same team

(Cláudio et al., 2015a). Although training and

assessing professional competence is a

multidimensional reality, which involves trainees’

work beyond passing or failing exams, but collecting

evidence of competence (i.e. the use of skills

according to standards of qualification) (Epstein and

Hundert, 2002), simulation is increasingly being used

for competencies based training and assessment. This

is a result of changes in health care delivery and

academic environments (e.g. the safety culture and

limited patient availability, respectively), and the

298

Cláudio, A., Carmo, M., Guerreiro, M., Cavaco, A., Pinto, V. and Pinha, A.

Virtual Humans Playing the Role of Patients in Self-medication Consultations: Perspectives of Undergraduate Pharmacy Students.

DOI: 10.5220/0005720002960303

In Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2016) - Volume 1: GRAPP, pages 298-305

ISBN: 978-989-758-175-5

shift to outcomes-based education, being clear the

advantages of simulation to demonstrate the

acquisition of basic competences (Scalese et al.,

2008). The present prototype simulates realistic self-

medication consultations using Virtual Humans

(VH), with the purpose of improving students’

clinical and communication skills in self-medication.

The prototype was developed with the aid of

Unity3D, a software tool for videogames. Both

Portuguese and English versions are available.

Firstly, users must select a self-medication

scenario and then, one VH must be chosen from four

characters (both genders and two different age

groups). The VH is depicted in a community

pharmacy environment and has natural body

movements, obtained by motion capture. Users,

playing the role of a pharmacist, communicate with

the VH by choosing textual options in the application

interface. Options are scored according to their level

of correctness at each point of the non-prescription

medicine consultation process; the goal is to obtain

the highest score by selecting the options that are

more correct.

The VH communicates verbally, by a synthetic

voice synchronised with the movements of the lips,

and non-verbally, by emulating facial and body

expressions as a response to the options chosen by the

user.

When the consultation is completed, users receive

feedback on their overall score and on their test score

in different consultation stages (e.g. patient

assessment and counselling). The application saves

the tests scores for each user, enabling data analysis

and progress monitoring over time.

A preliminary evaluation of the prototype was

conducted with seven experts (academic and

practitioners), from different organisations (Cláudio

et al., 2015a). This paper describes a user-study

involving two sets of students.

The paper is organized as follows: next section

presents the most relevant work in the area; sections

III to V describe our application; section VI reports

the evaluation; and section VII presents conclusions

and future work.

2 STATE OF THE ART

Healthcare education has used VH as a useful

resource (Cláudio et al., 2015b; Hoffman, 2000),

although its application in pharmacy education seems

scarce. Thus, one first step in ascertaining

pedagogical usefulness would be exploring the

advantages provided by VH in the development of

pharmacists’ clinical and communication skills.

A systematic review with meta-analysis on the use

of computer-based VH solutions for health

professions, evaluated the outcome of this

educational resource on users’ satisfaction,

knowledge, attitudes and communication/clinical

skills (Cook et al., 2010). Total of 3,285 learners

(2,115 medical students, 272 nursing students, both

professionals and other learners) allowed concluding

that VH were associated with large positive effects

compared with no intervention. However, VH effect

compared with non-computer instruction was on

average smaller, and not always statistically

significant, suggesting lower effectiveness on a wider

sample of healthcare learners. The large

inconsistency among study results led the authors to

exert caution when drawing conclusions about true

effectiveness differences between methods.

Cook and colleagues reported also results from

four qualitative studies with medical students, which

returned interesting accounts on learners’ views

concerning the use of VH (Cook et al., 2010). These

were perceived to bring advantages over other

educational methods, especially in what concerns

convenience of access, a relaxing learning

environment, and case evolution according to

learners’ actions and feedback. Actually, these were

key determinants of participants’ satisfaction and

engagement. Case realism was associated both with

the VH presentation and case material, including the

type of case presented. Additional topics contributing

to improve learning outcomes comprised repetition

until demonstration of proficiency, enhanced

feedback, and explicitly contrasting cases (Cook et

al., 2010).

A more recent review by Jabbur-Lopes and

colleagues focused on the use of VH in

pharmaceutical care education to pharmacy students

(Jabbur-Lopes et al., 2012). Using less stringent

criteria than the work of Cook et al., this review found

seven articles, four of which were conducted in North

America. The authors conclude that the paucity of

published studies involving VH suggests an under use

of this tool in pharmacy education and this conclusion

is supported by a cross-sectional survey in a

convenience sample of 194 European pharmacy

undergraduates (Cavaco and Madeira, 2012), where

only 12 students, from six universities in six different

countries reported experiences with VH. Those were

satisfied with this educational method and non-users

revealed themselves in favour of the potential benefits

of VH in pharmacy education.

Virtual Humans Playing the Role of Patients in Self-medication Consultations: Perspectives of Undergraduate Pharmacy Students

299

Computerised approaches make use of video clips

to create tools for skills training, while others fall

back on non-computerised virtual patients. One

example of the last comes from teaching self-care

consultation skills to pharmacy students in North

America (Orr, 2007); this study found an

improvement in overall knowledge and

communication skills. An example of the former i.e.

a VH application based on video images, photographs

of people in full size and realistic background sounds,

received the name of Virtual Practice Environment

(VPE) (Hussainy et al., 2012). This solution tries to

create a 3D immersive environment for high fidelity

patient care training in community pharmacies.

Furthermore, background images can also be

combined with previous recordings of staged

situations. Another example of high-end proposals is

the Interactive Simulated Patient (Bergin and Fors,

2003). This multimedia application was developed

for healthcare students, to explore and solve clinical

cases, aiming to help students’ practice of their

clinical reasoning skills. This is a sophisticated

resource where the interaction with the virtual patient

is made through natural language, with text input in a

dialog box, and the patient’s answers being offered

through video clips.

The AVATALK is a device that uses VH for skills

training in different areas (Hubal et al., 2000). The

main features comprise emotions expression with, the

possibility of behaviour modelling, using natural

language processing techniques. As with previous

proposals there is an option of integration of three-

dimensional scenarios. It was tested in medical

practice training together with another application,

the Trauma Patient Simulator, which defines

scenarios, medical histories, and the rules of

simulation between the patient and the trainee. The

learning results from the two last applications

evaluation were quite encouraging.

Compared to these solutions, our application has

additional features: it can be used both in training and

assessment mode; it includes a BackOffice

application to support teachers’ work, in particular it

offers an easy-to-use process for definition of new

self-medication cases that requires no informatics

skills from the user; and the application uses the

overall information contained in these definitions to

automate a set of choices concerning the VH that

populate the simulation. Next sessions gives details

about these features.

3 DESCRIPTION OF THE

APPLICATION

Virtual Pharmacy was designed to be used by

Pharmaceutical Sciences students in two usage

contexts: training and assessment. In both cases, the

student selects, from a list, a self-medication

situation. After that, he chooses a VH among a set of

four possibilities (both genres, young and old

characters).

The application recreates the chosen situation of

self-medication which takes place in a virtual

community pharmacy, represented as a background

image that is randomly selected from a set of images.

The VH plays the role of a patient suffering from that

particular clinical condition and requiring the

assistance of a healthcare professional (Figure 1). The

VH communicates verbally by a synthetic voice

synchronised with the lips.

Along with verbal communication, the VH also

communicates through body animations and by

changing his facial expressions (neutral, satisfied and

discontent). The initial expression is random and may

change in the course of the interaction, as a response

to the right or wrong choices made by the student

(Figure 2). The student, playing the role of a

pharmacist behind the counter, interacts with the VH

by selecting one of three textual options displayed in

the application interface (right side of Figure 1). Each

set of three options includes questions with different

levels of correctness to which are assigned different

scores. The course of the simulation is guided by the

options sequentially chosen by the student.

After performing a sequence of interactions with

the VH, the student completes the virtual consultation

and receives feedback on his overall scores.

When in training mode, the student uses the

application autonomously and, by the end of each

virtual consultation, together with his overall scores,

he receives detailed information about the correctness

of his previous choices and about his progress along

successive training sessions. The application also

provides links to educational contents related to the

topics of the self-medication situations presented.

In assessment mode, the application displays the

student’s overall scores and stores results in a

database for posterior analysis by the teacher with the

purpose of assigning him a mark.

GRAPP 2016 - International Conference on Computer Graphics Theory and Applications

300

Figure 1: Application interface- the animated virtual patient

describing his problem (a headache) using speech and

gesture; below this, the subtitles of the speech that

accompanies the speech sound; on the right side, the three

options presented for the user to choose one.

Figure 2: Virtual Patient exhibiting three distinct facial

expressions (from left to right): neutral, satisfied and

discontent.

3.1 Architecture of the Application

The Virtual Pharmacy application was developed for

PC. It has three main components: the VP_SIM, the

VP_Office and the Speech Generator. Each one of

these components contains several modules.

The VP_SIM is the core component that is

intended to be used by the students. It is implemented

in C#, using Unity3D, and comprises the User

Interface and the Application Controller that

commands the simulation of the situations.

The VP_Office implements the BackOffice

Application (BOA) that supports teachers’ tasks. Its

main component is the Dialogue Creator,

implemented using a javascript libray, JointJS

(jointjs.com), to build graphs. The Dialogue Creator

offers a graphical interactive interface to define or

edit self-medication situations, described as a graph.

Each node in the graph corresponds to a phrase that is

articulated by the VH or to an option and its

associated score. The nodes representing the speech

of the patient are connected to several nodes which

are the options to be shown to the student. In contrast,

the nodes that correspond to an option presented to

the student have only one link to a phrase of the

patient; this node corresponds to the reaction of the

virtual patient to the choice made by the student.

Using the graph to build the dialogues does not

require informatics skills.

The BOA application provides also other

functionalities for the teachers: an interface to upload

complementary elements of study to be shown in the

training mode of VP_SIM; and an interface to

monitor students’ progress over time. A MySQL

database is used to store the information which is

shared by the VP_SIM and the VP_Office.

When the VP_SIM application is launched, the

Data Controller module reads from the database the

information necessary to create the list of situations

to be displayed in the interface and to manage the

course of the simulations. Detailed information can

be obtained in (Cláudio et al., 2015a).

The third component, Speech Generator is

responsible for the VH correct articulation of the

speech as explained in the following section.

3.2 Animations and Speech

The four VH’s models in the application represent

adults of both genders and two different age groups:

young and old. They were obtained from DazStudio

(daz3d.com) and enriched with body and facial

animations using two different animation techniques:

skeletal animation for the body and morph target

animation for the face. Both facial and body

animations were combined in Blender (blender.org)

and then exported to Unity3D.

Facial animations comprise the expression of

emotions and the movement of the lips synchronized

with speech. Body animations were built based on

motion capture data downloaded from the Unity3D

assets store.

To attain a good level of realism, the

synchronization of the lips with the sound is an

essential feature in speech animation, because

observers are very sensitive to small errors (Xu,

2013). The component of the application that

processes speech generation is the Speech Generator

which relies on a Microsoft speech library

(www.microsoft.com/en-us/download/details.aspx

?id=10121) to convert text to speech. Taking into

account the age and gender of the character, for each

phrase of the dialogue two files are created: an audio

file, in wav format, with the speech of the character,

and an XML file, with the respective animation.

Body animations of the VH are triggered by

keywords in the dialogue that are connected to the

self- medication situation. For instance, the character

Virtual Humans Playing the Role of Patients in Self-medication Consultations: Perspectives of Undergraduate Pharmacy Students

301

brings his hand to the head while he articulates a

sentence describing the symptoms for headache. The

module Character Behavior Controller is responsible

for analysing the speech lines of the VH in the graph

to detect the keywords connected with animations.

This module also controls the facial animation of the

VH, reacting to the student’s performance while

using the tool.

4 EVALUATION

A preliminary version of VP_SIM was evaluated with

7 experts on pharmacy practice, to study the usability

and suitability of the application. The results of this

user-study revealed that these professionals

considered VP_SIM a valuable tool for training and

assessing students’ OTC (over-the-counter, i.e., drugs

that do not require a doctor’s prescription)

counselling skills (Cláudio et al., 2015a). But some

associated dangers were identified: normal

limitations related to a predefined dialogue that does

not cover all possible patient reactions and events in

a real situation. In their opinion, the contact with the

simulated situations in the application should

preferably be followed by real interactions in a true

community pharmacy.

The majority of panel experts were not supportive

of replacing the face-to-face teaching or the

internship, since there are behavioural competencies

that need real interactions and experience. In their

opinion, the application as a starting point should be

kept in a mix training environment. The subsequent

version of VP_SIM was fine-tuned and evaluated

with students.

4.1 Participants and Procedure

The evaluation resorted to two distinct sets of

volunteer Pharmaceutical Sciences students. A first

set comprised 10 undergraduate Pharmaceutical

Sciences students in their 4th year of studies; a second

set was composed by 42 students in the last year of

their Pharmaceutical Sciences course, performing

their curricular internship behind a counter in a

community pharmacy. Students in both sets were

from the same university to ensure identical learning

backgrounds. Each student tried the application,

performing the same list of suggested tasks and

afterwards answered to a set of questions in a Google

Forms questionnaire.

The list of suggested tasks included: i) to start the

application and make the registration and login steps;

ii) to solve the Headache situation choosing the

younger male character; iii) to solve the Cough

situation with a character of their own choice among

the available 4 in the interface; iv) and finally, to

consult the summarized information about his/her

performance in the resolution of the situations.

Besides the biographical questions to obtain a

general profile of the user, three groups of questions

were asked: i) specific questions about the interface

of the application and the VH, and some questions to

gather an overall evaluation; ii) a SUS (System

Usability Scale) questionnaire (Brooke,1996); and

iii) questions to infer about application’s suitability

and potential use for students’ training and

assessment in the area of Pharmaceutical Sciences.

The majority of the questions in groups 1 and 2

were answered using a Likert scale ( from 1- strongly

disagree to 5- strongly agree). All questions in group

3 were open-ended questions. Both set of users

answered to the 3 groups of questions in a Google

Forms questionnaire.

The user-study started with the 10 students (5

women, 5 men) in the 4

year. They made the tests in

a classroom in their own faculty, just after a class, and

all of them used the same computer with the

following characteristics: CPU- Intel Core i7, 2,4

GHz; RAM: 8GB; GPU: Nvidia GeForce GT 650M,

1GB.

The interviewer, observing the students while

they were performing the suggested tasks, was able to

perceive the difficulties the participants were

experiencing with the interface and registered all their

verbal comments. This feedback collected by the

interviewer was used to eliminate some ambiguities

in the interface of the application before the

evaluation with the internship students. The 42

internship students (35 women, 7 men) received an

email explaining how to download and execute the

application in their own computer, the list of

suggested tasks and the link to the Google Forms

questionnaire to fulfil afterwards.

Students of the 4th year, still with no contact

experience with real patients, used the training

version of VP_SIM and internship students used the

assessment version. As mentioned before, the latter

displays only the overall scores of the student, while

the training version provides detailed information

about the correctness of the previous choices, on the

progress over successive sessions of training and

links to educational content.

GRAPP 2016 - International Conference on Computer Graphics Theory and Applications

302

Table 1: Median values of the answers to questions about the interface (group 1).

Questions about the interface

(answered with a Likert scale: from

1- strongly disagree to 5- strongly agree)

4th year

students’

(median)

Internship

students’

(median)

The information is properly displayed on the screen (e.g. number and locations of

buttons/options)

4 4

2 Performing tasks is straightforward, i.e., the choice of options and/or menus is easy 4 4

3 The operation speed of the application is adequate 5 4

4 The choice of self-medication situations is clear 4 4

5 The choice of self-medication situations is straightforward, i.e., it does not require too many steps 4 4

6 The choices given to the user are meaningful 4.5 4

Table 2: Median values of the answers to questions about the Virtual Humans (group 1).

Questions about the Virtual Humans

(answered with a Likert scale: from

1- strongly disagree to 5- strongly agree)

4th year

students'

(median)

Internship

students’

(median)

1 The selection of the virtual human is clear 5 4

The selection of the virtual human is straightforward, i.e., it does not

require too many steps

5 4

3 The virtual humans are very realistic 4 3

4 The cloths of the virtual humans are very realistic 3.5 3

5 The way the virtual humans look is very realistic 4 3

6 The movements of the virtual humans’ bodies are very realistic 3 3

7 It is easy to hear / understand the sentences articulated by the virtual humans 4 4

8 The movements of the virtual humans’ lips are synchronized with the voice 4 3

The feedback provided by the virtual humans’ facial expressions is useful for the user of the

application

4 4

Table 3: Median values of the answers to questions about the overall opinion of the application (group 1).

Questions about the overall opinion of the application

(answered with a Likert scale: from

1- strongly disagree to 5- strongly agree)

4th year

students’

(median)

Internship

students’

(median)

I am content with the information the application gives me about my performance in the

resolution of the cases

5 4

I am content with the summarized information the application gives me about my performance

in the resolution of all the previous cases

5 4

3 The application is easy to use 5 4

4 The application responds to my anticipated goals 5 4

4.2 Results and Discussion

The results (median of users’ scores) of the questions

in group 1 are presented in table 1 (interface), table 2

(VH) and table 3 (general opinion). The overall

results are quite positive; lower scores appear in table

2 and are related to the realistic appearance and

behavior of the VH playing the role of patients.

Higher quality models, eventually commercial,

would most probably produce better user scores,

concerning the questions in lines 3, 4 and 5; body

movements can also be improved resorting, for

instance, to better motion capture data. The non-

controllable issue is in line 8, because a natural

speech articulation requires a very good

synchronization between lips’ animation and sound,

a demanding requirement in terms of hardware. It

should be noticed that internships used their own

computers, so we do not have control on the quality

of the hardware.

We could have asked those participants about the

hardware they were using. But we considered this

question counter-productive because i) that can be a

hard question to answer by someone potentially

without informatics skills, and ii) that question would,

most likely, prevent them from participating.

That is, a “less-good” computer may have

produced a lower impression on the user. This

speculation should be proved in future evaluations,

probably using always the same (type of) computer,

like 4

year students did. However, we have to be

cautious because this requirement of using the same

type of computer may hinder user testing.

Regarding the feedback provided by the virtual

humans’ facial expressions, reflecting his/her

performance (Table 2, line 9), students in the two sets

Virtual Humans Playing the Role of Patients in Self-medication Consultations: Perspectives of Undergraduate Pharmacy Students

303

agreed that it is useful (scores’ median = 4). Half of

the participants mentioned that more expressions

were needed, such as surprised, feeling pain,

doubtful, confused, concerned and angry.

Students in the 4th year scored with 5 the

questions in table 3 (general opinion), while

internships scored only with 4. Notice that these

results are not directly comparable because the last

ones tried the assessment version and some of them

wrote in the open-ended questions that they would

like to have more information about the resolution of

the cases, for instance the pathologies.

In the SUS questionnaire all the questions

received scores 4 and 5, except two that had a median

score of 3 and reveal that the users feel the usage of

the application is somehow restrictive. This result is

coherent with the mentioned small number of

available situations to solve, also expressed in open-

ended questions.

In the open-ended questions, 4

year students

emphasized as positive aspects the possibility of

interaction with the virtual patients resorting to a

logic dialogue that support the training of the

protocols (for instance, the correct sequence of

questions to pose); they also enjoyed the possibility

of choosing a particular character. The application

was considered as a valid learning tool, adequate to

help the learning process and to review the concepts

explained in classes.

As a negative characteristic they mentioned the

rapid end of the interaction with the patient when the

student makes a mistake at the beginning of the case

simulation.

The internships considered the prototype could

become a powerful tool to learn and to consolidate

knowledge, as long as it offers more situations. They

mentioned that it can be a tool for students, but also

for professionals that need to review and update their

professional capacities: it can help remembering

concepts and principles learned in school but already

forgotten.

As negative aspects, they referred that real

patients in pharmacy are usually not so predictable

and sometimes not as civilized as the virtual ones in

the application.

The majority of participants mentioned that a

negative aspect is the very limited number of

situations (just two, for the time being). All the

participants agreed that the tool cannot substitute a

real teacher, but it can be a good adjuvant learning

tool, corroborating the opinion of the experts that

tested the preliminary version of the prototype.

5 CONCLUSIONS AND FUTURE

WORK

We developed and implemented an interactive

application to support training and assessment of self-

medication consultation skills in pharmacy students,

using VHs animated with facial expressions, body

movements and speech communication.

The abbreviated usability study allowed checking

for potential advantages of using this software,

clearly as an additional source of students’ motivation

to practice their counselling skills scenarios: students

were optimistic in relation to move away from

traditional paper-based clinical cases into a closer

sense of professional interaction.

Another strong point of this prototype is its

BackOffice web application, which can be operated

by users without informatics skills. It allows

instructors to develop the consultation dialogues in an

intuitive and user-friendly graphical interface and

provides data on students’ performance.

The evaluation of pharmacy students was

positive, both in terms of usability and usefulness.

This is in accordance with experts’ views, who were

also of the opinion that the application could be used

for postgraduate continuing education.

The overall results obtained are encouraging and

justify the development of an improved version which

will respond to most users’ suggestions. Additionally

we intend to include principles of “gamification”, by

adjusting the level of difficulty to the scores

previously obtained. This may increase motivation of

young students to use the application in the training

mode. An escalation of difficulty relies, from a

communication standpoint, in more variability in the

VHs behaviour and in a wider range of simulated

moods (for instance, less cooperating patients, more

voice tones, more facial and body expressions).

Currently the application includes virtual humans

speaking Portuguese and English, a feature we wish

to maintain in the next version. In the near future, it is

also our intention to widen the evaluation to

pharmacy students and practitioners in Portuguese-

speaking countries. This will certainly shed light on

improvements and functionalities needed to yield

versions that are appropriate for different countries

and cultures.

ACKNOWLEDGEMENTS

We acknowledge the financial support of BioISI

R&D unit, UID/MULTI/04046/2013 funded by

GRAPP 2016 - International Conference on Computer Graphics Theory and Applications

304

FCT/MCTES/PIDDAC, Portugal and of the “Secção

Regional do Sul e Regiões Autónomas da Ordem dos

Farmacêuticos” that granted a BInov scholarship to

our research. We also thank Microsoft Portugal for

the license of the TTS voice.

REFERENCES

Asseray, N., Ballereau, F., Trombert-Paviot, B., Bouget, J.,

Foucher, N., Renaud, B., et al., 2013. Frequency and

severity of adverse drug reactions due to self-

medication: a cross-sectional multicentre survey in

emergency departments. Drug Safety, 36(12), pp.

1159–68.

Bergin, R. A., Fors, U. G., 2003. Interactive simulated

patient — an advanced tool for student- activated

learning in medicine and healthcare. Computers &

Education, 40(4), pp. 361–376.

Brooke, J., 1996. SUS-A quick and dirty usability

scale. Usability evaluation in industry, 189(194), pp. 4-

Cavaco, A. M., Madeira, F., 2012. European pharmacy

students' experience with virtual patient technology.

American J of Pharmaceutical Education, 76(6):

Article 106.

Cláudio, A. P., Carmo, M. B., Pinto, V., Cavaco, A.,

Guerreiro, M. P., 2015a. Virtual Humans for Training

and Assessment of Self-medication Consultation Skills

in Pharmacy Students. Proc. of IEEE 10th Int. Conf on

Comp Science & Education, Cambridge, pp. 175-180.

Cláudio, A. P., Carmo, M. B., Gaspar, A., & Teixeira, R.

2015b. Using Expressive and Talkative Virtual

Characters in Social Anxiety Disorder Treatment. Proc.

of 10th International Conf on Computer Graphics

Theory and Applic (GRAPP), Berlin, pp. 348-355.

Cook, D. A., Erwin, P. J., Triola, M. M., 2010.

Computerized virtual patients in health professions

education: a systematic review and meta-analysis.

Academic Medicine, 85(10), pp. 1589–1602.

Cooper, R. J., 2013. Over-the-counter medicine abuse: a

review of the literature. J of Substance Use: 18(2), pp.

82–107.

Donabedian, A., 1988. The quality of care: How can it be

assessed?. Jama, 260(12), pp. 1743-1748.

Epstein, R. M., Hundert, E. M., 2002. Defining and

assessing professional competence. Jama, 287(2), pp.

226-235.

Hoffman, H. M., 2000. Teaching and learning with virtual

reality. Studies in Health Tech. and Informatics, vol. 79,

pp. 285-291.

Hubal, R. C., Kizakevich, P. N., Guinn, C. I., Merino, K.

D., West, S. L., 2000. The Virtual Standardized Patient.

Medicine Meets Virtual Reality, pp. 133-138.

Hussainy, S. Y., Styles, K., Duncan, G., 2012. A Virtual

Practice Environment to Develop Communication

Skills in Pharmacy Students. American J of

Pharmaceutical Education, 76(10): Article 202.

Jabbur-Lopes, M. O., Mesquita, A. R., Silva, L. M., De

Almeida Neto, A. D. A., Lyra Jr., D. P, 2012. Virtual

patients in pharmacy education. American J of

Pharmaceutical Education, 76(5): Article 92.

King, S. A., Parent, R. E., 2005. Creating Speech-

Synchronized Animation. IEEE Trans on Visualization

and Computer Graphics, 11(3), pp. 341-352.

Martins, A. P., Miranda, A. D. C., Mendes Z., Soares M.

A., Ferreira, P., Nogueira, A., 2002. Self-medication in

a Portuguese urban population: a prevalence study.

Pharmacoepidemiol Drug Safety, 11(5), pp. 409–14.

Orr, K. K., 2007. Integrating Virtual Patients into a Self-

Care Course. American J of Pharmaceutical Education,

71(2): Article 30.

Pereira, D. V., Cavaco, A. M., 2014. Exploring computer

simulation to assess counseling skills amongst

pharmacy undergraduates. Indian J of Pharmaceutical

Education & Research, 48(1), pp. 17-26.

Scalese, R. J., Obeso, V. T., Issenberg, S. B., 2008.

Simulation technology for skills training and

competency assessment in medical education. J of

General Internal Medicine, 23, Suppl 1, pp. 46-49.

Serra, J., Ribeiro, M., Freitas, J., Orvalho, V., Sales Dias,

M.S., 2012. A Proposal for a Visual Speech Animation

System for European Portuguese. In IberSPEECH

2012, CCIS 328, pp. 267-276.

Schmiedl S, Rottenkolber M, Hasford J, Rottenkolber D,

Farker K, Drewelow B, et al., 2014. Self-medication

with over-the-counter and prescribed drugs causing

adverse-drug-reaction-related hospital admissions:

results of a prospective, long-term multi-centre study.

Drug Safety, 37(4), pp. 225–35.

Xu, Y., Feng, A. W., Marsella, S., Shapiro, A., 2013. A

Practical and Configurable Lip Sync Method for

Games. MIG '13, ACM, New York, 2013, Article 109.

Virtual Humans Playing the Role of Patients in Self-medication Consultations: Perspectives of Undergraduate Pharmacy Students

305