Patient Education and Counseling 96 (2014) 381–388

The educational value of improvisational actors to teachcommunication and relational skills: Perspectives ofinterprofessional learners, faculty, and actors

Sigall K. Bell a,b,*, Robert Pascucci c, Kristina Fancy d, Kelliann Coleman e,David Zurakowski f, Elaine C. Meyer g

a Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, USAb Institute for Professionalism and Ethical Practice, Boston Children’s Hospital, Harvard Medical School, Boston, USAc Critical Care Medicine and Perioperative Anesthesia, Departments of Critical Care Medicine and Anesthesia,

Boston Children’s Hospital, Harvard Medical School, Boston, USAd Institute for Professionalism and Ethical Practice, Boston Children’s Hospital, Harvard Medical School, Boston, USAe Institute for Professionalism and Ethical Practice, Boston Children’s Hospital, Harvard Medical School, Boston, USAf Department of Anesthesia and Director of Biostatistics for the Departments of Anesthesia and Surgery,

Boston Children’s Hospital, Harvard Medical School, Boston, USAg Department of Psychiatry, and Director, Institute for Professionalism and Ethical Practice,

Both at Boston Children’s Hospital, Harvard Medical School, Boston, USA

A R T I C L E I N F O

Keywords:

Communication

Simulation

Education

Interprofessional

Actors

A B S T R A C T

Objective: To assess the educational value of improvisational actors in difficult conversation simulations

to teach communication and relational skills to interprofessional learners.

Methods: Surveys of 192 interprofessional health care professionals, and 33 teaching faculty, and semi-

structured interviews of 10 actors. Descriptive statistics, Fisher’s exact test and chi-square test were used

for quantitative analyses, and the Crabtree and Miller approach was used for qualitative analyses.

Results: 191/192 (99.5%) interprofessional learners (L), and 31/33 (94%) teaching faculty (F) responded

to surveys. All 10/10 actors completed interviews. Nearly all participants found the actors realistic (98%L,

96%F), and valuable to the learning (97%L, 100%F). Most felt that role-play with another clinician would

not have been as valuable as learning with actors (80%L, 97%F). There were no statistically significant

differences in perceived value between learners who participated in the simulations (47%) versus those

who observed (53%), or between doctors, nurses, or psychosocial professionals. Qualitative assessment

yielded five actor value themes: Realism, Actor Feedback, Layperson Perspective, Depth of Emotion, and

Role of Improvisation in Education. Actors independently identified similar themes as goals of their

work.

Conclusions: The value attributed to actors was nearly universal among interprofessional learners and

faculty, and independent of enactment participation versus observation. Authenticity, feedback from

actors, patient/family perspectives, emotion, and improvisation were key educational elements.

� 2014 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

Patient Education and Counseling

jo ur n al h o mep ag e: w ww .e lsev ier . co m / loc ate /p ated u co u

‘‘Giving the actors the freedom to respond in an improvisa-tional way mirrors the life experience of these very importantmedical conversations. Even when a medical team goes intoa team meeting with a prepared ‘script’ for the conversationor what they decide ahead of time what they want to cover

* Corresponding author at: Beth Israel Deaconess Medical Center, 330 Brookline

Ave; CO-1309; Boston, MA 02215, USA.

E-mail address: Sbell1@bidmc.harvard.edu (S.K. Bell).

http://dx.doi.org/10.1016/j.pec.2014.07.001

0738-3991/� 2014 Elsevier Ireland Ltd. All rights reserved.

or accomplish, these scenarios with actors . . . reminds themedical team that these conversations should never be‘scripted.’ There is not a set formula for doing theseconversations. The actors’ authenticity allows this teachingpoint to resonate. . . and invites the medical participants toenter into a learning environment that emphasizes that the artof communication and relationship building is an ongoing,living process.’’

–A faculty member

S.K. Bell et al. / Patient Education and Counseling 96 (2014) 381–388382

1. Introduction

Communication with patients and families in today’s healthcare environment is increasingly complex. Whether focused ondisclosure of medical error, prenatal diagnoses made possible bynew technologies, or ethical quandaries arising from the ability tosustain life, today’s clinicians routinely face difficult conversationsas part of day-to-day patient care. Effective communication istherefore increasingly a critical component to the delivery ofquality care. It can also build trust, improve clinical outcomes, anddecrease litigation [1–5].

Recognizing the relational complexities inherent to theseconversations, medical educators are emphasizing principledcommunication skills, to build trust and partner with patients[6–10]. They are also increasingly turning to role-play, simulatedor standardized patients, and other such experiential teachingmodalities to practice difficult conversations in a safe learningspace [11–13]. While learners are taught strategies and evenprotocols to approach challenging conversations, the reality isthat no two clinical conversations are the same. A successfulapproach in one setting may be inadequate or even problematicwith another patient or family. Relational proficiency inchallenging conversations necessitates recognition of, andflexible response to, variable patient and family cues. Educatorsare stymied not only by finding a way to teach this relationalflexibility, but also how to evaluate it. They also face thechallenge of successfully bridging lessons from instructionalsettings to clinical practice [8].

Several approaches have been taken to the simulated patient’srole in education and research, often focusing on standardizedpatients [12,14], but relatively less is known about theexperiences of simulated patients in such educational interven-tions, [15] or the perceived value of improvisational actors toteach communication and relational skills from the integratedperspectives of learners, faculty, and actors themselves. Althoughstandardized patients and actors have been used interchange-ably, in this report we distinguish between the two. Whilestandardized patients are traditionally guided to provide a‘‘standard’’ performance – the same each time – actors are highlyimprovisational. By definition, no two conversations are thesame.

We developed the Program to Enhance Relational Communica-tion Skills (PERCS) in response to the growing need forinterprofessional educational programs centered on challenginghealth care conversations [13,16–19]. A central component of thePERCS approach is the use of live enactments with professionalactors to portray patients and family members using improvisa-tional techniques.

The purpose of this study was to examine the value of actors inthis educational paradigm from the perspectives of interprofes-sional learners, teaching faculty and the actors themselves.

2. Methods

2.1. Participants and educational workshops

Since 2002, the Program to Enhance Relational and Commu-nication Skills (PERCS) has trained more than 3000 local, nationaland international learners. The workshops, which includeenactments with professional actors portraying the role ofpatients and family members, range from 4 to 8 h, and involveapproximately 15–25 interprofessional learners gathered aroundcommon ethical communicative challenges. Learners includedoctors, nurses, medical interpreters, and psychosocial profes-sionals (e.g., psychologists, chaplains, child life specialists)identified as part of the healthcare delivery team. Workshops

are facilitated by physician, psychosocial, and family faculty, andaddress a range of challenging conversations in neonatology,neurology, surgery, pediatrics, medicine, radiology, anesthesia,and critical care.

The enacted cases are fully vetted for clinical relevance andaccuracy by PERCS content experts and family faculty, includingphysician, nurse, and social work experts in ethics, end-of-life care,medical error disclosure, spirituality, and communication; as wellas families with extensive medical needs. The cases are tailored tothe clinical department participating in the workshop, but oftenfocus on conveying bad news, navigating ethical dilemmas withpatients and families, or medical error disclosure. CharacterDevelopment Tools are prepared for professional actors to providerealistic ‘‘back-stories’’ that create deep factual, emotional, andsocial foundations from which to ground improvisation [20]. Afterreading through the scenario as a group, volunteer learnerparticipants join actors in live, improvisational enactments.Non-volunteering participants view enactments live or throughclosed circuit video. Each enactment is immediately followed by aformal debriefing led by facilitators and including the actors,enactment participants, and other learners in reflective discussion.Further details of the PERCS pedagogy have been previouslydescribed [13,17–19].

2.2. Evaluation of actors and educational paradigm

We asked all 192 learners from 13 medical and surgicalworkshops conducted from October 2010 to June 2011 to completepre- and post-workshop questionnaires. The questionnaires weredeveloped and vetted by a research team including communica-tion skills experts, physicians, ethicists, and a statistician. Theyincluded both Likert-style questions and open-ended narrativequeries, examining participant attitudes and comfort related todifficult conversations, as well as assessment of learning method-ologies used in the workshop, and required 10–15 min tocomplete. All 33 PERCS teaching faculty were also asked tocomplete questionnaires of similar structure, and ten PERCSactors were invited to participate in semi-structured interviewsabout their experiences in the workshops. The interviewsfollowed a structured guide including 10 questions that werealso developed by the research team to probe actor experiencesrelated to their roles in the educational sessions. Forty to sixtyminute telephone, or e-mail (1 actor) interviews were conductedby one research assistant after the questions had been vetted bythe PERCS leadership team. The interviews were then transcribedverbatim.

Participation by workshop learners, faculty members, andactors was voluntary. The Institutional Review Board of BostonChildren’s Hospital determined that PERCS met educationalexemption criteria.

2.3. Quantitative analysis

We used descriptive statistics to analyze learner and facultycharacteristics. Pearson chi-square (x2-test) was used to evaluatedifferences in responses between the four studied professionalgroups (doctors, nurses, psychosocial professionals, and medicalinterpreters). Fisher’s exact test was used to compare responses ofenactment participants and enactment observers. In addition,multivariable logistic regression was used to assess whetherparticipant perceptions about effectiveness of role-play versusactors was associated with age, gender, ethnicity, experience,discipline, or enactment status. Statistical analyses were per-formed using SPSS software (version 19.0, SPSS Inc./IBM, Chicago,IL). Two-tailed values of P < 0.05 were considered statisticallysignificant.

Table 1Learner characteristics.

Characteristic (n = 191)

Age, mean � SD, years 37.6 � 10.7

Years of clinical experience since earning a

professional degree, median (range)

7 (1–42)

Gender

Female 158 (83%)

Male 33 (17%)

Ethnicity

Caucasian 152 (80%)

Hispanic 22 (11%)

Asian 8 (4%)

African American 5 (3%)

Other 4 (2%)

Profession

Nurse 114 (60%)

Physician 43 (22%)

Psychosocial professional* 20 (11%)

Medical interpreter 14 (7%)

Area of specialty

Pediatric critical care 74 (39%)

Surgery 67 (35%)

Neonatal critical care 30 (16%)

Adult and Pediatric Anesthesia 20 (10%)

SD = standard deviation.* Includes social workers, psychologists, chaplains, child life specialists.

S.K. Bell et al. / Patient Education and Counseling 96 (2014) 381–388 383

2.4. Qualitative analysis

Open-ended survey responses to the question ‘‘What specifi-cally was valuable about the actors’ participation in the scenariosand debriefing?’’ were reviewed by 4 members of the study team.Two investigators (RP, KC) used the Crabtree and Miller model forthematic analysis of learner responses; 2 different investigators(EM, SB) conducted thematic analysis for faculty responses usingthe same model [21]. Each investigator independently reviewed allthe qualitative narratives and compiled a list of key themes derivedfrom the participant responses. Each group then met to comparetheir respective list of themes in a reconciliation process with theirpartner, where assigned themes and their interpretations wereadjudicated through in-depth conversation to resolve any differ-ences of opinion. After establishing consensus about themes, eachgroup created their own coding manual. Both groups then returnedto the data, assigning themes to qualitative responses to test forsaturation of themes. The two groups then came together tocompare their findings in a face-to-face meeting, and final themesreflecting both learner and faculty responses were establishedthrough consensus. At least one team member was present at themajority of the workshops, serving as a ‘‘member check’’ for theface validity of the emerging themes.

We interviewed actors about their acting experience, and theirroles in the PERCS workshops. One investigator (SB) reviewed actorinterview transcripts to elicit responses to 5 structured interviewquestions.

3. Results

3.1. Participants

Of all workshop learners, 191/192 (99.5%) completed ques-tionnaires and 188/191 (98.4%) submitted both pre- and post-surveys. The average participant age was 37.6 � 10.7 with a medianof 7 years of clinical experience (range 1–42). Of the learners, 60%were nurses, 22% were physicians, 11% were psychosocial profes-sionals and 7% were medical interpreters (Table 1). We receivedsurvey responses from 31/33 (94%) PERCS faculty members, and 10/10 (100%) of actors.

3.2. Expectations and experience

3.2.1. Learners

Most learners had optimistic expectations for actors as aneducational strategy in the pre-workshop assessment: 75% werepositive, 17% were neutral, and 8% were skeptical. After theworkshops, the vast majority of learners felt the clinical scenariosin the enactments were realistic (99%) and clinically useful (96%).Nearly all (98%) described the actors’ portrayal of the patient/family as realistic, and (97%) agreed that the participation of theactors was valuable to their learning about communication skillsfor challenging conversations with patients and families (Table 2).

The majority of learners (80%) stated that role-play withanother clinician colleague would not have been as educationallyvaluable as learning with improvisational actors (Fig. 1). First,

Table 2Learner perceptions of actor-based pedagogy stratified by professional training.

Educational Outcome All (n = 191) Nurses (n = 114) Physician

Scenario realistic 189 (99%) 113 (99%) 42 (98%)

Scenario clinically useful 183 (96%) 110 (97%) 40 (93%)

Actors realistic 188 (98%) 113 (99%) 42 (98%)

Actors valuable 185 (97%) 112 (98%) 39 (91%)

Role-play not as effective as actors 152 (80%) 92 (82%) 35 (81%)

learners found that the actors offered a ‘‘fresh, unbiased perspec-tive.’’ They reported that ‘‘unfamiliar faces make it more realistic,’’and that actors would not ‘‘be nice’’ or ‘‘make [the conversation]easier’’ for them because of underlying friendships. Second,involving actors in the teaching demanded attention to medicaljargon and relating to patients as lay people, concepts that mightbe missed if practicing exclusively with healthcare colleagues.Third, learners questioned whether role-play with colleaguescould ‘‘convey the intensity of emotion,’’ accessed by actors.

Multivariable logistic regression confirmed that role-play wasnot as effective as actors, and was independent of age, gender,ethnicity, and years of experience. There were no significantdifferences between professions (P = 0.15) or between enactmentparticipants compared to observers (P = 0.46).

Of all learners, 47% participated in the live enactments, and 53%observed them. There were no statistically significant differencesbetween enactment participants versus observers for the five mainoutcomes (learner perceptions of realistic scenarios, clinicallyuseful scenarios, realistic portrayal by actors, educational value ofactors, and role play not as effective as actors, Table 3). Similarly,there were no significant differences in outcome measures whencomparing learners of different professional backgrounds(Table 2).

3.2.2. Faculty

Teaching faculty reported mixed expectations about the actors’contribution to the educational experience: 40% were positive, 30%were neutral, and 30% were skeptical. Skeptical respondentsworried that the enactments would not be realistic, and felthesitant to trust the actors. Faculty thought the actors would be

s (n = 43) Psychosocial (n = 20) Interpreters (n = 14) P value (x2-test)

20 (100%) 14 (100%) 0.78

20 (100%) 13 (93%) 0.54

19 (95%) 14 (100%) 0.52

20 (100%) 14 (100%) 0.07

17 (85%) 8 (57%) 0.15

Rol

e-Pl

ay N

ot a

s Ef

fect

ive

as A

ctor

s (%

)

0

10

20

30

40

50

60

70

80

90

100

83%78%

81%82% 85%

57%

ProfessionEnac tment Status

“Role-p lay not as effec�ve ”

Fig. 1. Percentage of learners who believed role-play would not have been as

effective as improvisational actors, stratified by professional training and

enactment status (participant versus observer).

S.K. Bell et al. / Patient Education and Counseling 96 (2014) 381–388384

stilted and unconvincing in their roles, and expected to be criticalof the actors’ performance. Neutral respondents expected a‘‘standardized patient approach, with relatively fixed and definedgoals and agendas.’’

After leading workshops with improvisational actors, 97% offaculty agreed the actors’ portrayal of the scenario was realistic,and 100% felt that the actors’ contribution was valuable to thelearning. Nearly all (94%) faculty felt that the actors’ improvisationcontributed to the authenticity of the scenarios. At the same time,61% felt that the unpredictability inherent to improvisation wasanxiety-provoking for faculty, reflecting lack of control about whatthe actor would precisely say during the enactment and thedebriefing. However, 94% of faculty reported that actors had soundemotional rationale for their feedback. Nearly all (97%) of thefaculty members stated that role-play would not have been aseducationally valuable as simulation with improvisational actors.

3.3. Most valuable contribution of actors

We asked learners and faculty what they found most valuableabout the actors’ participation. Of all participants completingsurveys 167/191 (87%) learners and 29/31 (94%) teaching facultyprovided narrative responses, yielding 5 central themes.

3.3.1. Realism

Learners and faculty commonly described the deep authenticityof actors in their scenario roles. As one faculty member noted,

Table 3Learner perceptions of actor-based pedagogy stratified by enactment participation.

Educational outcome Enactment

participants

(n = 90)

Enactment

observers

(n = 101)

P value

(Fisher’s

exact

test)

Scenario realistic 89 (99%) 100 (99%) 1.00

Scenario clinically useful 89 (99%) 94 (93%) 0.07

Actors realistic 89 (99%) 99 (98%) 1.00

Actors valuable 87 (97%) 98 (97%) 1.00

Role-play not as effective as actors 70 (78%) 82 (83%) 0.46

‘‘The actors allow caregivers an opportunity to have [thedifficult situation] as close to real life as it can get. It feels veryreal.’’

Learners noted that actors ‘‘acted so well it was easy to suspenddisbelief,’’ and that this quality made them take the training moreseriously. One added that the realism ‘‘increases [my] confidenceto handle similar situations.’’

3.3.2. Importance of actor feedback

Learners frequently remarked, ‘‘It was helpful to hear . . . exactlywhat they felt, heard, and how much they could handle,’’ notingthat they often wonder how the patient and/or family feels inresponse to their communication but generally cannot ask inroutine clinical practice. Faculty agreed:

‘‘The ability to ask them (actors) how did it feel when theclinician did or said something. No simulator can tell you how itfelt.’’

Faculty also reflected on their own willingness to ‘‘shareexpertise’’ with actors as teachers, noting that actors often delivercustomized teaching points powerfully.

‘‘I think it was amazing that an actor would state, ‘I felt veryuncomfortable when you told me ___.’ It did not occur to methat the actors would be able to actually feel and then expresssuch emotions. To be honest, I was blown away by theirinsights.’’

3.3.3. The actor’s layperson perspective

Actors’ lack of medical knowledge was cited as an asset andimportant to the learning experience. Recognizing that ‘‘. . .[clin-[clinicians] do not usually know how the information is interpretedor even heard,’’ learners ascribed value to understanding what laypeople take away from interactions with health care providersThey noted paying extra attention to avoiding medical jargon, andadded:

‘‘[Their] outside perspective, not from a medical background,allowed them to be more realistic and [raise] questions andviewpoints as a family generally would.’’

Faculty also noted that actors had no previous or futurerelationship with learners, and from this ‘‘unattached’’ vantagepoint, could both speak up (in a way patients relying on healthprofessionals may fear doing) and provide honest feedback (incontrast, for example, to a colleague who the learner might see orwork with clinically):

‘‘Actors have the ability and freedom to say what patients andfamilies cannot say, [and] to be very honest in a safeenvironment.’’

3.3.4. Depth of emotion

Learners and faculty noted that the ability of actors to ‘‘dial upor down’’ particular emotions effectively optimized learning andpushed learners who were at different levels. One learner noted:

‘‘They are outstanding proxies for parents in difficult situations;they can combine the emotional and spiritual truth of themoment with a safe offering of its reflection to the partici-pants.’’

Another added:

‘‘The actors frame their teaching in the emotional moment ofthe enactment, not through a subjective judgment from outsideit.’’

and Counseling 96 (2014) 381–388 385

3.3.5. Role of improvisation in education

Learners valued the actors’ ability to adapt flexibly to theirapproach to the conversation, noting that it ‘‘Feels very real, tohave to ‘think’ on your feet as in real life.’’ They also commentedon how actors redirected the discussion when necessary toensure that difficult topics were not avoided. Faculty recognizedthe actors’ improvisational skills as not only keeping eachenactment unscripted and unpredictable, but also enabling themto meet learners ‘‘where they are’’ or at the edge of theirlearning:

‘‘[Actors] capture the participants so completely that they losetheir ‘defenses’ and become themselves, opening up theirhidden talents in a away they didn’t realize could happen. Laterin the debriefing, actors help participants reflect on andunderstand what truly happened in the room.’’

Faculty perceived actor improvisation as a key component tothe PERCS pedagogy, underscoring the role of ‘‘reflection in action’’for both actors and learners. Faculty felt that this approachaccessed the unpredictable—calling on learners to ‘‘adapt andrespond in the moment,’’ and ‘‘allow[ing] for the surprises of realpractice.’’

3.4. Actor interviews

3.4.1. Actor experience

The actors had worked with the PERCS program for a mean 5.3years (range 2–9). Half (50%) had done prior medical educationwork—predominantly as standardized patients (SP). Asked tocomment on the difference between the PERCS improvisationalapproach and the SP role, they noted that the SP work was ‘‘veryfocused on the medicine,’’ and that ‘‘It was very important that I didthe same thing each time.’’ One actor noted:

‘‘You just needed to know the statistics of your condition andwhat medications you were on. It wasn’t really performing.’’

In contrast, actors suggested, ‘‘PERCS gets into the thing a littlebit deeper,’’ and ‘‘delves into questions of an ethical nature,’’Another added:

‘‘[PERCS] concentrates much more on the communication skillsaspect of medicine and also the kind of more ‘human’ aspects,the way people relate to each other.’’

3.4.2. Realism

Asked what makes the scenarios real, actors mostfrequently cited calling on their own personal experienceas patients or family members. They also highlighted theimportance of preparation, and emotional commitment to therole, commenting:

‘‘Be prepared to invest with your own life. It’s a real in depthform of improvisation.’’

3.4.3. Actor skills

Interviewees identified four key skills for successful enactmentsin health care education:

Listening and improvisation. Actors frequently referred toflexibility and the ability to ‘‘think on your feet.’’ They alsodescribed ‘‘knowing when to push [learners],’’ and to ‘‘expect theunexpected’’ and be ‘‘open minded’’ since learners may approachthe scenario in very different ways.

Full character immersion. Actors routinely felt that their ownfull ‘‘buy in’’ was important for realistic portrayal of characters, andthat they needed to be fully acquainted not only with theircharacter’s history, but also their ‘‘emotional situation.’’

S.K. Bell et al. / Patient Education

Teacher/feedback role. Actors routinely emphasized their roleas teacher, especially in giving feedback. Several noted that theexperience is ‘‘50/50’’ acting and providing feedback. Theydescribed a sense of responsibility for taking care of the learner,treating them with respect and kindness. The actors aimed to givefeedback in a nonjudgmental manner that ‘‘might make them try itdifferently next time.’’

Concentration/attention to detail. Actors noted the impor-tance of responding with full emotional commitment in themoment, while also taking note of specific phrases used, bodylanguage, tone, and other characteristics of the encounter tolater be integrated when providing feedback to workshopparticipants.

3.4.4. Debriefing/feedback with learners

Asked what they contributed to the debriefing, actors oftenlinked the value of their ‘‘firsthand feedback’’ to their role as alayperson ‘‘stand in’’ for the patient or family.

‘‘It’s kind of like having a split awareness, a dual awareness. . .

you have to be there really in the moment that it’s happening,but also just be aware of what’s impacting on you, like if theysay something that’s particularly helpful, or if they saysomething that’s particularly hurtful or insensitive or makesyou confused.’’

Actors also viewed their realism and feedback as a way toremind conditioned clinicians of the gravity of some medicalconversations.

‘‘I feel a lot of times doctors and nurses see this level of, they seethese medical problems very, very frequently, so they getdesensitized to what it’s like. I think a lot of times it’s constantlywhat we all learn in the room.’’

Like learners and faculty, actors viewed the debriefing as a keycomponent to the pedagogy.

‘‘You’ve got someone who can honestly portray that situationbut that is also looking as sort of a third eye, paying attention,and is then able to give feedback, so I think that is what we[actors] contribute to the program.’’

4. Discussion and conclusions

4.1. Discussion

Our study findings suggest that improvisational actors can havean important role in communication and relational skills training.For both learners and faculty, the value attributed to actors wasnearly universal (97–100%), and independent of direct participa-tion versus observation in the enactment – a finding that haspotential implications regarding scalability of training efforts.Similarly, the perceived value of actors was also independent ofprofessional backgrounds – including doctors, nurses, medicalinterpreters, and psychosocial professionals.

Learners and faculty themes about the educational valueattributed to actors suggest that each group specifically valuedthose qualities that actors aimed to accomplish in their work:realism, thoughtful feedback focused on the patient and familyexperience, a layperson perspective, depth of emotion, andimprovisation-driven reflection (Table 4). Actor feedback washighlighted as a key component to the learning, and can help fill anexisting gap in communication and relational skills training [22].Actors viewed themselves as a ‘‘third eye’’ – observing theconversation even as they participated in it, and then laterreflecting with the group on the encounter. In this way, theirprocess reflects the ‘‘think aloud’’ format [15]. Studies suggest

Table 4‘‘What is most valuable about actors?’’ Qualitative themes and quotes of learners, faculty, and actors.

Theme Learners Faculty Actors

Realism ‘‘Live portrayals help us to more

realistically bridge the gap between

theory and practice’’

‘‘[The actors are] skillful enough that

participants can suspend reality and

genuinely enter the experience in a way

that can be more real than real’’

‘‘We do our homework about who are

these people [patients or family members

we are portraying], and how are they

related to each other, and what is the

background’’

Importance of Actor Feedback ‘‘Getting feedback from the actors about

what it felt like to be on the family side

of the conversation. We can never ask a

family’s opinion of how situations are

handled or what parts were pivotal to

them

‘‘I find that the actors are actually better

at providing constructive feedback than

the facilitators. [The] experienced

clinicians actually listen more to the

constructive feedback when it is coming

from the actors’’

‘‘There’s really two parts to the PERCS stuff,

... one half is the scenario where you are

acting out the scene, whatever it is, the

other half is the debriefing where you go

back and talk about it, and that’s almost as

important as the scenario itself’’

‘‘I think [actors) are able to say things to a

clinician, to whoever is in the scenario with

them, that, that clinician might not hear

from their peers. You have just shared this

interaction and so they are more likely to

hear it coming from us’’

The Layperson Perspective ‘‘Incorporating actors is what makes

this program. It is so useful to have

professionals that can act, ad lib, and do

not have a medical background – it is so

realistic and helpful to see

conversations unfold’’

‘‘[Actors serve] as ‘ethical understudies’

for patients; they are used beautifully

give a voice to the patient and family

perspective’’

‘‘As outsiders to the medical system, we are

more like patients and we are able to see

things from a different perspective because

we are not in it every day; I think that is

critical to the whole process’’

Depth of Emotion ‘‘They conveyed a level of emotion that

might have been missing if one of us

just volunteered’’

‘‘It gave the emotional, visceral

response that is present in real

interactions’’

‘‘We are taught to be vulnerable to access

our emotions, to get inside the characters,

the skin of whoever we are playing. . . if you

use that as your starting point, then the

way in which you are interacting with the

clinician is going to be genuine’’

Role of Improvisation in

Education

‘‘Feels very real, have to ‘‘think’’ on your

feet as in real life’’

‘‘There have been many moments in

which the (unscripted) words or actions

of an actor makes my heart stop as an

educator’’

‘‘This is not scripted, you do not know what

you are going to get from the participants

that come in and you need to adjust and

react honestly’’

‘‘Some doctors and nurses can be pushed,

and need to have, so we’ll take their

scenarios to a different level if we are not

getting the right sort of information from

them’’

S.K. Bell et al. / Patient Education and Counseling 96 (2014) 381–388386

feedback from simulated patients increases learners’ motivationallevels and self-efficacy [23].

Improvisation and unscripted interactions mirror real practice,and are central aspects of the actor-based pedagogy. Actors canaccelerate learning with skillful use of questions such as, ‘‘Whatdoes that mean?’’ preventing learners from glossing over or shyingaway from difficult topics. Actors’ improvisational abilities enablethem to interact with all members of the health care team in thesame conversation, challenge interprofessional dynamics, infusepalpable emotion into difficult conversations, respond in ways thatmay be unanticipated by learners, and display their owninterpersonal conflicts with actor-family members, thereby moreaccurately reflecting the nuance and unpredictability of the clinicalenvironment. In addition, while other studies have noted aninversion of power dynamics in conversations involving SPs andmedical students [24] noting that institutional learning settingsmay ‘‘empower’’ SPs in ways that are not typical of the patient-doctor relationship [19,25], participation by multiple teammembers of varying levels of training makes the power dynamicmore closely aligned with actual practice.

As with any new learning experience, we expected that the‘‘buy in’’ from learners might be limited, especially for those whoassigned less value to improving communication with patients andfamilies or already felt proficient with such skills. We weresurprised to learn that faculty initially had more reservations aboutthe actor-based pedagogy than learners did, and that regardless ofany pre-existing concerns, the great majority (95–99%) of bothlearners and faculty felt the enactments were realistic andclinically useful, and that the actors were valuable to the learning.

What are the potential downsides of actor-based improvisa-tional learning? One ‘‘cost’’ is that by definition, faculty have lesscontrol over how a given enacted conversation will unfold (ascompared to, for example, a standardized patient encounter)—asituation that can lead to some anxiety on the part of faculty, as weobserved in this study. Despite this, faculty demonstrated strongenthusiasm for the use of actors and reported trusting theimprovisational process and talent of the actors. The literal pricetag of actor-based training is also not trivial. While the cost of anygiven workshop can vary considerably depending on length,number of participants, and number of scenarios; the actor feesand time for workshop planning are necessary expenses [20].

Why should educators invest in this pedagogical approach? Forall the reasons described by learners and faculty in this study, role-play with colleagues may not be an adequate substitute forimprovisational actors in teaching the requisite skills for challeng-ing conversations. Earlier studies cite no difference between use ofstandardized patients (SPs) compared to peers to learn motiva-tional interview techniques [26–28]. The strong preference foractors evident in our study may reflect the contrast betweeneducational approaches of improvisational actors compared to SPs(more standardized). In addition, the content studied in the latterreports (basic interview skills or smoking cessation approaches)are far less dependent on the deep emotional responses accessedby actors in life-threatening scenarios, and therefore, perhaps notsurprisingly, were similar between SPs and peers.

We suggest that an improvisational actor-based pedagogy isespecially useful in settings where emotion is a strong part of theconversational landscape, and understanding the patient/family

S.K. Bell et al. / Patient Education and Counseling 96 (2014) 381–388 387

perspective is especially critical. Actors are also uniquely suited toscenarios in which improvisation enlightens rote approaches;when the target communication skills are to be used with patientsor family members of a demographic not represented in colleague–peer groups (such as adolescents [29], bereaved parents, elders)but are easily accessed by actors; and when objective feedback thatis not constrained by a pre-existing or future relationship with thelearner – a concept we call ‘‘the actor’s freedom’’ – is particularlyvital. Under these circumstances, actors may offer a particularadvantage over more standardized patients or role play withcolleagues.

Given the expense, we recommend improvisational actorsimulation be reserved for more challenging healthcare conversa-tions. For example, a circumscribed focus on physical exam skills ina standardized Observed Structured Clinical Examination (OSCE)setting does not necessarily call for improvisational actors.Similarly, more basic interview skills may also be taught withstandardized patients or peers [26–28]. However, learning todiscuss end-of-life care options with parents of pediatric patients,disclosing a medical error to a harmed patient or his/her familymembers, or navigating abnormal prenatal testing results, asexamples, may greatly benefit from the use of improvisationalactors.

Limitations of this study include reliance on self-reportmeasures. The learners and faculty, although representing severalteaching hospitals, were based at one academic center, and theirattitudes may not be reflective of individuals at other centers.Nonetheless, the nearly universal appreciation of the pedagogy byour participants of varied departments, units, and disciplines(often each with their own local culture) suggests that theapproach can engage learners of many backgrounds. Furtherresearch focusing on long-term educational outcomes, such asperformance on communication skills assessments, and patient-provider relationships is needed.

4.2. Conclusions

Improvisational enactments mirror actual practice, and arecentral aspects of the value of actor-based improvisationalpedagogy. Interprofessional learners and teaching faculty iden-tified realism, direct feedback from actors, providing the patient/family perspectives, emotional depth, and the role of improvisa-tion as key educational elements, each of which mapped directlyto actors’ specific skills, intentions, and priorities in their work.The value attributed to actors was nearly universal amonginterprofessional learners and faculty, and independent oflearners’ professional discipline, or direct participation versusobservation in the enactment. Actor-based simulation offers arobust, engaging educational strategy to teach communicationand relational skills.

4.3. Practice implications

Should educators adopt actor improvisation for interprofes-sional communication and relational skills training at their owninstitutions? The results of our study highlight several insightswe have gained from a decade of teaching and learning withimprovisational actors. First, we humbly respect the power ofthis pedagogy to successfully ‘‘put practice before us.’’ Likepatient care, each workshop is decidedly different, each withunique teachable moments. As improvisational learningapproaches gain attention [30–32], teaching faculty also needto hone reflection-in-action to identify these critical momentsand learning opportunities.

Second, preparation matters. We have found that a deliberateprocess of actor auditions, training and character development,

including the use of a formalized Character Development Tool [20],have been extremely effective to address pertinent medical factsand to adequately develop the emotional and relational back-ground from which the actors can then improvise effectively.

Third, we have been struck by the finding, as described in thisstudy, that actors are as valuable to medical education for whatthey lack – a medical background, or a personal history or futurewith the learner (enabling more honest feedback) – as they are forwhat they bring: deep palpable emotion, realism, and a bridge forreflection.

Fourth, through our interprofessional workshops we areroutinely awed by how little most clinicians know of theirinterprofessional colleagues, and how much they can learn fromone another – both experts and novices – when given the chance.[13,33] Teams that work together may greatly benefit fromlearning together. Health care education desperately needsfrequent and robust opportunities for interprofessional learning[34], and actors can serve as a ‘‘common ground,’’ linking themembers of each group through their portrayal of the patient andfamily experience, and challenging team member dynamics in‘‘real time.’’

Fifth, after watching hundreds of conversations unfold over theyears with improvisational actors, we have come to appreciate thatwhile we teach communication strategies and approaches, the reallearning about communication and relational skills lies in theunscripted. This uncertainty can be unsettling to educators, asdescribed in this study, but like theater or jazz, improvisation isgrounded in rules that can be taught [35]. Regarding communica-tion and relational training, we believe clinicians need principledguidelines and a toolbox, rather than a checklist, and educatorsneed more flexible metrics with which to assess learners. To matchthe complex demands of effective communication in modern dayhealth care, we will need ever more creative learning environ-ments and evaluation processes. Improvisational actors can openthe door to one such approach—simultaneously challenging andengaging learners and faculty in unscripted clinical encounters,and providing a deeper space for reflection, moral imagination, andfeedback from patient and family perspectives.

Finally, health care professionals have traditionally beenreluctant or conservative in embracing the emotional tenor ofclinical care. All too often, clinicians may try to avoid patients whenthey anticipate uncomfortable discussions. We rarely acknowl-edge, talk of, or teach about, the courage medicine requires, thewishes we hold for our ailing patients, and the grief we experiencewhen things do not go well [36–39]. Perhaps actors – experts inemotion – can help us deconstruct the complexities of suchfeelings, and then bring them forward in ways that can promotewhole providers steadied to treat whole patients.

Disclosures

None for all authors.

Previous presentations

Portions of this paper were presented in part at theInternational Pediatric Simulation Symposia and Workshops,New York, NY. April 2013, and the International Conference onCommunication in Healthcare, Sept 2013, Montreal, Canada.

Acknowledgments

The authors thank David M. Browning MSW, BCD, Robert D.Truog MD, Elizabeth A. Rider MSW, MD, Allyson McCrary, BA,Diane M. Gentile, BS, and all the PERCS participants and faculty fortheir contributions to this work. SKB thanks the Arnold P. Gold

S.K. Bell et al. / Patient Education and Counseling 96 (2014) 381–388388

Foundation for a career development award through a GoldProfessorship.

References

[1] Lown BA, Rosen J, Marttila J. An agenda for improving compassionate care: asurvey shows about half of patients say such care is missing. Health Aff(Millwood) 2011;30:1772–8.

[2] Fiscella K, Meldrum S, Franks P, et al. Patient trust: is it related to patient-centered behavior of primary care physicians? Med Care 2004;42:1049–55.

[3] Boulding W, Glickman SW, Manary MP, Schulman KA, Staelin R. Relationshipbetween patient satisfaction with inpatient care and hospital readmissionwithin 30 days. Am J Manag Care 2011;17:41–8.

[4] Kachalia A, Kaufman SR, Boothman R, et al. Liability claims and costs beforeand after implementation of a medical error disclosure program. Ann InternMed 2010;153:213–21.

[5] Levinson W, Roter DL, Mullooly JP, Dull VT, Frankel RM. Physician-patientcommunication. The relationship with malpractice claims among primary carephysicians and surgeons. J Am Med Assoc 1997;277:553–9.

[6] Silverman JD, Draper J, Kurtz SM. The inhumanity of medicine. Interpersonaland communication skills can be taught. Br Med J 1995;310:527.

[7] Draper J, Silverman J, Hibble A, Berrington RM, Kurtz SM. The East AngliaDeanery Communication Skills Teaching Project-six years on. Med Teach2002;24:294–8.

[8] Kurtz S, Silverman J, Benson J, Draper J. Marrying content and process inclinical method teaching: enhancing the Calgary–Cambridge guides. Acad Med2003;78:802–9.

[9] Kurtz SM, Silverman J, Draper J. Teaching and learning communication skills inmedicine. 2nd ed. London: Radcliffe Pub.; 2005.

[10] Silverman J, Kurtz SM, Draper J. Skills for communicating with patients.London: Radcliffe Pub.; 2005.

[11] Dudley F. The simulated patient handbook: a comprehensive guide for facil-itators and stimulated patients. London: Radcliffe Publishing; 2012.

[12] Bokken L, Rethans JJ, Scherpbier AJ, van der Vleuten CP. Strengths and weak-nesses of simulated and real patients in the teaching of skills to medicalstudents: a review. Simul Healthc 2008;3:161–9.

[13] Meyer EC, Sellers DE, Browning DM, McGuffie K, Solomon MZ, Truog RD.Difficult conversations: improving communication skills and relational abili-ties in health care. Pediatr Crit Care Med 2009;10:352–9.

[14] Howley L, Szauter K, Perkowski L, Clifton M, McNaughton N. Quality ofstandardised patient research reports in the medical education literature:review and recommendations. Med Educ 2008;42:350–8.

[15] Lingemann K, Campbell T, Lingemann C, Holzer H, Breckwoldt J. The simulatedpatient’s view on teaching: results from a think aloud study. Acad Med2012;87:179–84.

[16] Waisel DB, Lamiani G, Sandrock NJ, Pascucci R, Truog RD, Meyer EC. Anesthe-siology trainees face ethical, practical, and relational challenges in obtaininginformed consent. Anesthesiology 2009;110:480–6.

[17] Lamiani G, Meyer EC, Leone D, et al. Cross-cultural adaptation of an innovativeapproach to learning about difficult conversations in healthcare. Med Teach2011;33:e57–64.

[18] Meyer EC, Brodsky D, Hansen AR, Lamiani G, Sellers DE, Browning DM. Aninterdisciplinary, family-focused approach to relational learning in neonatalintensive care. J Perinatol 2011;31:212–9.

[19] Browning DM, Meyer EC, Truog RD, Solomon MZ. Difficult conversations inhealth care: cultivating relational learning to address the hidden curriculum.Acad Med 2007;82:905–13.

[20] Pascucci RC, Weinstock PH, O’Connor BE, Fancy KM, Meyer EC. Integratingactors into a simulation program: a primer. Simul Healthc 2013;9:120–6.

[21] Crabtree BF, Miller WL, editors. Doing qualitative research. 2nd ed., ThousandOaks, London, New Delhi: Sage Publications, Inc.; 1999.

[22] Bokken L, Linssen T, Scherpbier A, van der Vleuten C, Rethans JJ. Feedback bysimulated patients in undergraduate medical education: a systematic reviewof the literature. Med Educ 2009;43:202–10.

[23] Turan S, Uner S, Elcin M. The impact of standardized patients’ feedback on thestudents’ motivational levels. Proc Soc Behav Sci 2008;1:9–11.

[24] de la Croix A, Skelton J. The simulation game: an analysis of interactionsbetween students and simulated patients. Med Educ 2013;47:49–58.

[25] de la Croix A, Skelton J. The reality of role-play: interruptions and amount oftalk in simulated consultations. Med Educ 2009;43:695–703.

[26] Papadakis MA, Croughan-Minihane M, Fromm LJ, Wilkie HA, Ernster VL. Acomparison of two methods to teach smoking-cessation techniques to medicalstudents. Acad Med 1997;72:725–7.

[27] Mounsey AL, Bovbjerg V, White L, Gazewood J. Do students develop bettermotivational interviewing skills through role-play with standardised patientsor with student colleagues? Med Educ 2006;40:775–80.

[28] Lane C, Hood K, Rollnick S. Teaching motivational interviewing: usingrole play is as effective as using simulated patients. Med Educ 2008;42:637–44.

[29] Hardoff D, Schonmann S. Training physicians in communication skills withadolescents using teenage actors as simulated patients. Med Educ 2001;35:206–10.

[30] Watson K. Perspective: serious play: teaching medical skills with improvisa-tional theater techniques. Acad Med 2011;86:1260–5.

[31] Khoury M, Saab BR, Haidar L. Autonomy and beneficence in an interactivetheater. Fam Med 2010;42:476–8.

[32] Case GA, Brauner DJ. Perspective: the doctor as performer: a proposal forchange based on a performance studies paradigm. Acad Med 2010;85:159–63.

[33] Kneebone R. Perspective: simulation and transformational change: the para-dox of expertise. Acad Med 2009;84:954–7.

[34] Hoffman SJ, Rosenfield D, Gilbert JH, Oandasan IF. Student leadership ininterprofessional education: benefits, challenges and implications for educa-tors, researchers and policymakers. Med Educ 2008;42:654–61.

[35] Yanow D. Learning in and from improvising: lessons from theater for organi-zational learning. Reflections: society for organizational learning and theMassachusetts Institute of Technology, vol. 2(4). 2011;p. 58–62.

[36] Gaufberg E. On courage. Acad Med 2010;85:805.[37] Carter BS, Brown JB, Brown S, Meyer EC. Four wishes for Aubrey. J Perinatol

2012;32:10–4.[38] Delbanco T, Bell SK. Guilty, afraid, and alone—struggling with medical error. N

Engl J Med 2007;357:1682–3.[39] Browning DM, Meyer EC, Brodsky D, Truog RD. Reflections on love, fear, and

specializing in the impossible. J Clin Ethics 2007;18:373–6.