G Model

PEC-4894; No. of Pages 12 Patient Education and Counseling xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou

Review

Doctor–patient communication skills training in mainland China: A systematic review of the literature Xinchun Liu a,b, Wesley Rohrer c, Aijing Luo b,*, Zhou Fang d, TianHua He e,f, Wenzhao Xie b a

Department of Social Medicine & Health Management, School of Public Health, Central South University, Changsha, China Third Xiangya Hospital of Central South University, Changsha, China c Graduate School of Public Health, University of Pittsburgh, Pittsburgh, USA d Swanson School of Engineering, University of Pittsburgh, USA e School of Medicine, TsingHua University, Beijing, China f School of Medicine, University of Pittsburgh, USA b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 7 November 2013 Received in revised form 11 September 2014 Accepted 20 September 2014

Objectives: To conduct a systematic review of studies on doctor–patient communication skills training (CST) for medical students and physicians in mainland China. Methods: We retrieved articles from six electronic databases, and searched additional eligible papers by checking reference lists. Chinese or English-language studies focused on CST and implemented in mainland China were applied to the pre-determined criteria. Articles included were further reviewed under the following categories: participant; training strategy; assessment; and outcome. Results: 20 studies met the inclusion criteria. 90% of the CST improved trainees’ communication skills using a strategy which included a didactic component combined with practical rehearsal and feedback. The duration of training varied substantially. A lack of enhancement in empathy, and the use of openended questions were reported. 83% of the assessment instruments were self-designed and most lacked reliability and validity testing. Only two of the included studies evaluated patient satisfaction. Conclusions: The majority of included studies attained statistically significant improvements. Chinese doctors and medical students’ communication skills can be enhanced through CST. Practice implications: Future studies in China should place stronger emphasis on the development of training strategies, validation of the assessment instruments, and evaluation of patient satisfaction affected by CST. ß 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: Doctor–patient communication skills Training Physician Medical student China Systematic review

Contents 1. 2.

3.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data source and searches . . . . . . . . . . . . . . . . . . 2.1. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Data extraction and quality assessment . . . . . . 2.3. Data synthesis and analysis . . . . . . . . . . . . . . . . 2.4. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. General characteristics of the included studies . Training strategies . . . . . . . . . . . . . . . . . . . . . . . 3.2. Training content . . . . . . . . . . . . . . . . . 3.2.1. Training methods. . . . . . . . . . . . . . . . . 3.2.2. Duration of training. . . . . . . . . . . . . . . 3.2.3.

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

000 000 000 000 000 000 000 000 000 000 000 000

* Corresponding author at: Third Xiangya Hospital of Central South University, Tongzipo Road 138, Changsha, Hunan 410013, China. Tel.: +86 731 886 186 67; fax: +86 731 888 619 10. E-mail address: [email protected] (A. Luo). http://dx.doi.org/10.1016/j.pec.2014.09.012 0738-3991/ß 2014 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 2

X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

3.3.

4.

Assessment instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chinese version of scales or questionnaires adapted from Western countries . 3.3.1. Self-developed scales or questionnaires, examinations and narrative reports . 3.3.2. Effects of communication skills training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Reported changes in trainees’ communication skills . . . . . . . . . . . . . . . . . . . . . 3.4.1. Trainees’ self-confidence in communication skills improvement . . . . . . . . . . . 3.4.2. Trainees’ attitudes toward training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.3. Patient satisfaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.4. Discussion and conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. 4.3. Implications for research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction The significance of doctor–patient communication has been confirmed by evidence-based research, and shows effective doctor–patient communication relating to outcomes such as enhanced patient satisfaction, better treatment compliance and symptom resolution [1–5]. Deficiencies in communication are associated with negative patient experience, medical errors, and malpractice litigation [6–8]. Analysis of complaints regarding physicians’ behaviors indicates that the most common categories reported by consumers relate to communication and interpersonal skills [9]. Communication skills involve the ability to convey information to another effectively and efficiently. Good communication combines a set of skills including nonverbal communication, attentive listening, the ability to build up trust and respect, emotion handling, and shared-decision making. Communication is not a skill that develops automatically over time and with experience [10], many studies have confirmed that communication skills can be taught and the mastered competence can be retained over years [11–14]. Doctor–patient communication in China has gained increasing attention as the relationship between patients and health care providers has sharply deteriorated over the past decade [15]. Investigations have indicated that 98.47% of hospitals have experienced medical disputes involving patient complaints [16]. Violence against medical professionals has become a common phenomenon; doctors and nurses have been murdered by patients or their family members [17,18]. The reasons of this deterioration include, but are not limited to, any of the following: defects in health policy and regulation, deficiency in humane quality, information asymmetry, poor doctor–patient communication, and physician’s overloaded performance pressure [19,24]. According to a national survey, more than two-thirds (67.7%) of the whole sample (N = 4875) agreed that there are tensions and conflicts between doctors and patients; and slightly more than 70% of the medical respondents stated that inadequate communication with patients prevented improvement in the doctor–patient relationship [60]. Consequently, this situation demands practical measures to improve physicians’ caring competences and interaction skills with patients. The theories and methodologies of doctor–patient communication skills training (CST) were introduced in China almost 10 years ago. Western CST models such as the Calgary– Cambridge Guide [63] and the SEGUE Framework [50] have been adapted with modifications based on Chinese culture. Nanjing Medical University developed and deployed the first doctor– patient communication curriculum in 2003 [20]. However, recent statistics demonstrated that only 40% of Chinese medical schools include doctor–patient communication in their curriculum, and most of the CST was an optional course [21]. Furthermore, CST continuing education for residents and practicing physicians is rare

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

000 000 000 000 000 000 000 000 000 000 000 000 000 000

in Chinese hospitals [61]. Compared to the identified substantial need for doctor–patient CST, the resources and endeavor devoted to it are, at best, inadequate [22]. The majority of published articles on doctor–patient communication have focused on theoretical arguments such as the importance of CST, root cause analysis of ineffective communication, and recommendations for better communication [20]. Few studies have focused on the implementation and effectiveness of CST. With regard to the evaluation of CST effectiveness, even though the Chinese version of some well-established CST assessment tools such as the SEGUE scale [23] and the Liverpool Communication Skills Assessment Scale (LCSAS) have been verified [62], the application of the main body of measurements for doctor–patient CST is unclear. To our knowledge, no systematic review has been conducted on doctor–patient CST as applied to Chinese medical students and practice physicians. This review has been undertaken to address the following questions: (a) What training strategies are being applied to implement CST in mainland China? (b) What assessment instruments are used to evaluate CST in mainland China and do they have proven reliability and validity? (c) How effectively is CST applied in mainland China? (d) What recommendations can be provided to improve research and best practices in CST in mainland China? 2. Methods 2.1. Data source and searches We searched the following databases: CNKI, WANFANG, PubMed, Embase, Ovid, Proquest ARL, PsychInfo, Communication and Mass Media, and Social Sciences Citation Index. The first two are the largest and most commonly used databases for retrieving Chinese academic articles and dissertations. All studies indexed in the searched databases as of August 1, 2013 were potentially eligible. The search strategy in all the databases used combinations of Keywords and MeSH terms found to be relevant from pilot searches. The keyword combinations used were the following: ‘doctor–patient communication’ or ‘physician–patient communication’ or ‘doctor–patient relations’ or ‘physician–patient relations’; ‘physician’ or ‘doctor’; ‘medical student’; ‘train*’ or ‘medical education’; and ‘China’. A further combination of ‘communication AND physician patient relationship AND China’ was added as a MeSH term to narrow the search in these databases. Reference lists of selected articles were examined to identify additional potentially eligible articles. 2.2. Study selection We included only studies of doctor–patient CST with an evaluation of effectiveness. We defined doctors as physicians or

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

medical students including residents, general practitioners, specialists, and interns. Medical students also included student psychologists as the specialty of most medical students in China is ‘‘clinical medicine’’, however, some students focus on ‘‘clinical psychology’’. As the curriculum design in ‘‘clinical psychology’’ is similar to that in ‘‘clinical medicine’’ during the first two to three college years, these students may attend the same doctor– patient CST courses. If there was no specific description in the included articles, we did not distinguish between them. Considering the differences in training contents and outcome measurements, we did not include studies on CST targeted at nurses and medical technicians. To be included in the review, the study had to meet the following inclusion criteria: (a) either randomized controlled trials (RCTs), controlled clinical trials (CCTs), pre–post test designs (PPTDs) or post-test designs (PTDs); (b) focus on CST; (c) trainees were solely or mainly medical students or doctors; (d) assessment tools such as scales, questionnaires or narrative reports were applied and clearly described; (e) relevant outcomes were assessed, but were not limited to any of the following: trainees’ communication skills competence, trainee’s self-confidence in communication skills improvement, patient satisfaction, and patient health outcomes. We excluded studies which: (a) were not implemented in mainland China and not reported in Chinese or English; (b) included fewer than 10 participants. 2.3. Data extraction and quality assessment Two authors (WZX and THH) who were responsible for the database searches reviewed the article titles to exclude those which were clearly ineligible and the primary author (XCL) reviewed the abstracts of the remaining articles to identify studies that were potentially eligible. Three authors (XCL, ZF and THH) reviewed the full text of the potentially eligible articles. Two authors (XCL and WZX) checked the reference lists of the selected full-text papers and conducted a follow-up review of the additional articles for potential inclusion. Disagreements regarding study inclusion were resolved by discussion with the senior authors (AJL and WR). We contacted one author by phone for clarification regarding the outcome analysis of the study, and contacted the Chinese Medical Doctor Association office for clarification on the participants. Based on the Cochrane Collaboration [25] and PRISMA guidelines [26], we evaluated the risk of bias in each study by examining specific study components relevant to this topic and constructed a ‘‘risk of bias’’ table (Table 3). We evaluated major sources of bias in intervention research [27], such as participant selection bias, performance bias, attrition bias, and measurement bias. A high risk of results bias could not be excluded if there was no clarification on participants, or the number of participants evaluated was small. For example, a study [28] synthesized participants from China, Lao and Vietnam; however, no specific results were obtained for the 79 Chinese participants, only 47 of a total of 200 participants were assessed and only 3 participants were interviewed after the CST. Even though it was a valuable investigation of CST in Chinese physicians, we decided to exclude this study based on the exclusion criteria.

communication skills competence, trainee’s self-confidence in communication skills improvement, trainees’ attitudes toward training, and patient satisfaction or health outcome as affected by the interaction with physicians who received CST. 3. Results 3.1. General characteristics of the included studies We retrieved 1527 articles from our database searches. Of the initial sample, 546 duplicate articles were removed and 1314 were excluded after reviewing article titles and abstracts. After reviewing the remaining 213 papers in full text, 23 articles were added based on a reference review. Fifty-nine articles were retained for further screening and 20 articles (18 in Chinese and 2 in English) were finally included in the review after applying the inclusion and exclusion criteria. Fig. 1 summarizes the study selection process. The included studies (2 RCTs, 4 CCTs, 10 PPTDs, and 4 PTDs) are described in Table 1. A total of 3309 participants were included in the 20 studies. Almost 99.3% (n = 3285) of the participants were physicians, interns, and medical students; the remaining 0.7% (n = 24) were nurses and patients who had also received CST. Of these studies, one was developed by the cooperation between U.S. universities

1527 citations identified: CNKI (633) WANFANG (398) Pubmed (125) EBASE (51) Ovid (56) Proquest ARL (115) PsychInfo (114) SSCI (26) Communication & Mass Media (9)

546 duplicate articles removed

981 Titles reviewed

528 ineligible articles removed

453 abstracts reviewed 240 articles excluded: Theoretical elaboration on CS (223) Majority of trainees are not physicians or medical students (17)

213 full-text articles reviewed 23 articles added after reference reviewing 177 articles excluded: No description of CST intervention (144) Not in mainland China (21) Trainees were nurses (12)

59 papers retained for further screening

2.4. Data synthesis and analysis For the included studies, we evaluated all communication skills training processes including participant selection, training strategy, the application of assessment instruments, statistical analysis and summary or composite measures of CST. We classified studies according to the research design and outcomes assessed. Four measures of CST were evaluated: changes in trainees’

3

20 papers included after applying both

39 articles excluded: No assessment of CST effect (28) Assessment of CST effect not clearly described (9) Participants not clarified (1) Participants < 10 (1)

inclusion and exclusion criteria Fig. 1. Flow chart of study selection.

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

4

Table 1 Study characteristics (studies are organized in chronological order within study design). Study

Study design

Participants

Training methods

Duration of training

Theoretical background

Outcome

Yu et al. (2010) [30]

RCT

Harris’s and Escovitz’s reports of using SSP in CST

RCT

R: SSP + ESP C: traditional teaching R: SP + CST C: SP

Unclear

Zhang et al. (2010) [31]

228 Interns (R = 114; C = 114) 46 Interns (R = 22; C = 24)

Unclear

Hang et al. (2009) [40]

CCT

160 Interns (R = 82; C = 78)

Unclear

Trainee’s CS competence Trainee’s self confidence

Sun et al. (2011) [34]

CCT

99 Medical students (R = 47; C = 52)

Unclear

Barrows’ PBL teaching theory

Trainee’s self confidence Trainee’s attitude toward CST

Li et al. (2011) [35]

CCT

100 Medical students (R = 50; C = 50)

1 session

Unclear

Trainee’s CS competence

He et al. (2013) [43]

CCT

234 Interns (R = 123; C = 111)

R: lecture; video watch; case study; role-playing; clinical interview C: traditional internship R: PBL; scenario simulation C: traditional teaching R: role-playing; feedback C: traditional teaching R: lecture; SP C: traditional internship

2 sessions/ week; 48 sessions in total (6 months) Unclear

Trainee’s CS competence Trainee’s attitude toward CST Trainee’s CS competence

Unclear

HMPS model

Trainee’s CS competence Patient satisfaction

Pan et al. (2009) [36]

PPTD

38 Ophthalmologists residents

Lecture; role-playing; scenario simulation; practice analysis

3 sessions (2 weeks in total)

Kaiser

Lai et al. (2009) [44]

PPTD

40 sessions in 5 days

HMPS model

Wu et al. (2010) [42]

PPTD

61 Residents

20 sessions in 5 days

HMPS model

Jian et al. (2010) [29]

PPTD

52 General practitioners

Kalamazoo Census Statement

Wang et al. (2010) [47]

PPTD

28 General practitioners

8 sessions Follow-up: 2 months Unclear

Trainee’s Trainee’s CST Trainee’s Trainee’s Trainee’s CST Trainee’s

Unclear

Trainee’s self-confidence

Jiang (2011) [41]

PPTD

82 General practitioners

1 year

HMPS model SEGUE framework

Trainee’s CS competence

Sun et al. (2012) [39]

PPTD

23 Physicians

10 sessions

Trainee’s self-confidence Trainee’s attitude toward CST

Wang et al. (2012) [48]

PPTD

61 Interns

SPIKES model Informed consent guideline for cancer patients SEGUE framework

Chen et al. (2013) [32]

PPTD

30 General practitioner residents

SEGUE framework

Trainee’s CS competence

Wuensch et al. (2013) [38]

PPTD

SPIKES model

Trainee’s self-confidence

Blatt et al. (2009) [45]

PTD

24 Oncologists and 7 oncology nurses 49 Medical professionals (most were physicians)

5 days (40 sessions)

PCC model (Later developed into HMPS model by CMDA)

Trainee’s self-confidence Trainee’s attitude toward CST

Lai et al. (2010) [37]

PTD

642 Medical students

16 sessions

Unclear

Shi et al. (2010) [33]

PTD

52 Medical students

Unclear

Textbook ‘‘Doctor–patient communication’’ developed by Nanjing medical college

Trainee’s Trainee’s CST Trainee’s Trainee’s CST

Permanente’s Four Habits Model 1263 Physicians and interns

Trainee’s CS competence Patient satisfaction Lecture; case discussion; video watch; role-playing Lecture; case discussion; roleplaying; focus group SP; feedback; focus group Lecture; roleplaying; video analysis; feedback SP; role-playing; lecture; focus group; video feedback Lecture; roleplaying; small group discussion Lecture; roleplaying; clinical interviewing SP; lecture; one to one coaching; internship

Role-playing; group discussion; video watch Small group teaching: lecture; discussion; video feedback; roleplaying Clinic internship; SP; case study; role-playing SP&SFM; scenario simulation; case discussion; internship

Unclear

Lecture: 2 sessions SP interview: 2 sessions Internship: 4 weeks 1.5 day (12 sessions)

self confidence attitude toward CS competence self confidence attitude toward CS competence

Trainee’s CS competence Trainee’s self-confidence

self-confidence attitude toward CS competence attitude toward

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

5

Table 1 (Continued ) Study

Study design

Participants

Training methods

Duration of training

Theoretical background

Outcome

Zhu et al. (2011) [46]

PTD

13 physicians, 3 nurses, 14 patients

Case discussion; role-playing; experience sharing

Unclear

Unclear

Trainee’s CS competence Trainee’s attitude toward CST

CST: communication skills training; CS: communication skills; R: research group; C: control group; RCT: randomized controlled trial; M: male; F: female; CCT: controlled (non-randomized) clinical trials; PPTD: pre–post test design; PTD: post-test design; SP: standardized patient; SFM: standardized family member; SSP: student standardized patient; ESP: electronic standardized patient; HMPS: humanistic medicine practice skill; SPIKES: six-steps of good communication (Setting, patients Perception, Information need, provide Knowledge, responding to emotions with Empathy, Summary and strategy); PCC: patient-centered communication; CMDA: Chinese medical doctor association.

and the Chinese Medical Doctor Association (CMDA) [45]. As this study defined the trainees as ‘medical professionals’, for verification, we obtained information from the CMDA to confirm that physicians were the main participants. One study was conducted by German experts [38], in which Chinese oncologists were trained on breaking bad news to patients with cancer. 3.2. Training strategies 3.2.1. Training content The training contents of the 20 included studies were divided into two categories as follows: 3.2.1.1. Communication skills training models adapted from Western countries. One study [29] used training contents adapted from the Kalamazoo Consensus [49] and two [32,48] used the SEGUE Framework [50], which addressed a comprehensive range of communication skills including initiating the session, gathering information, physical examination, explanation and planning, and closing the session. The Four Habits Model (FHM) [51] was modified for the Chinese context in four domains including invest in the beginning, elicit patient’s perspective, demonstrate empathy, and invest in the end [36]. Two studies [38,39] focused on improving trainees’ communication skills in breaking bad news to patients with cancer, adapting SPIKES [52] (Setting, patient Perception, Information need, provide Knowledge, responding to emotions with Empathy, Summary and strategy) as the primary training model. Patient-centered communication (PCC) was introduced to Chinese doctors in a train-the-trainer program [45]. 3.2.1.2. Communication skills training models developed by Chinese institutions or project teams. Four studies [41–44] employed a humanistic medicine practice skill (HMPS) model – a curriculum developed by the Chinese Doctor Association (CDA) with the support of U.S. universities [45]. The HMPS consists of five modules: ethics/professionalism, relationship building, information gathering, explaining the problem and agreeing on a plan, and breaking bad news. By training both healthcare providers and patients, one study [46] placed emphasis on medication adherence when communicating with patients with HIV/AIDS. Handling medical conflicts and coping with work stress were added in response to patients’ mistrust and violence against medical professionals in China [33,40]. Another six studies [30,31,34,35, 37,47] developed their own training content based on principles of doctor–patient communication skills. 3.2.2. Training methods Communication skills training in the included studies were mostly characterized by a combination of cognitive and experiential elements. This entailed both taught communication theoretical knowledge and practiced communication skills. The interventions made use of a variety of training methods including standardized patient, standardized family member, role-playing, scenario simulation, feedback, group discussion, and lectures. Most of the CST employed multiple training methods during the intervention.

Standardized patients (SP) were applied in seven studies [29,31–33,37,41,43], and student standardized patient (SSP) in one study [30]. Another study [33] employed standardized family members (SFM) into the training. This approach aimed to mitigate the increasingly severe doctor–patient conflicts in China. Roleplaying was used in 13 studies [35–42,44–48], and scenario simulation in 3 studies [33,34,36]. These approaches were conducted under controlled conditions, in the presence of colleagues and experienced facilitators. Feedback was given to the trainees based on direct observation or a review of the recorded encounters in five studies [29,35,41,45,47]. Among the studies targeted at physicians or interns, group discussion was their priority [29,33,37–42,44–46]; and specific cases of disease or medical disputes were discussed and perspectives were elicited. Lectures were used in most of CST sessions, which provided the participants with a theoretical background of doctor–patient communication. 3.2.3. Duration of training The duration of CST varied substantially. Studies on the HMPS model included a fixed period of training: about 20–40 sessions within 5 days [42,44,45]. Two studies delivered 10 [38] and 12 [39] sessions, respectively, focusing on skills useful for breaking bad news to patients with cancer. Duration of training for medical students or physicians with SP ranged from 4 weeks [32] to 1 year [41]. Several studies [30,33,34,40,43,46–48] did not describe the exact duration of training. We managed to calculate the number of sessions involved in short-term communication training, which was found to be a mean of 17 sessions (range 1–40, standard deviation 14.35) over 2.5–3 days. 3.3. Assessment instruments The effects of CST were evaluated using two types of measurements as follows: except for the adapted Chinese versions of SEGUE, Four Habits Coding Scheme (4HCS) and Breaking bad news Assessment Schedule (BAS), the validity and reliability of other assessment instruments applied in the included studies had not been tested. 3.3.1. Chinese version of scales or questionnaires adapted from Western countries Among the included studies, SEGUE was the most commonly used scale [32,41,48] for evaluating the effects of CST. SEGUE was translated into Chinese and verified in 2008, with 5 domains and 25 items [23]. One study [29] used a scale adapted from the Kalamazoo Assessment Tool [49], with a Chinese version of 5 modified domains and 20 items based on the current doctor– patient communication situation in China. The Four Habits Coding Scheme (4HCS) [53] was translated into Chinese, with the 4 original domains modified to 5 domains and the original 26 items modified to 32 items [36]. The Breaking bad news Assessment Schedule (BAS) [54] was adapted and verified in one study [39], with 12 items adapted from the original 23 items. The domains of the above scales were closely related to the domains of the training

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

6 Table 2 Results categorized by category of outcomes. Outcome

Study/training model or method

Assessment instrument

Domain measured

Item and score

Comments on results

Trainee’s CS competence (studies = 11)

Jiang et al. [29] (SP)

Kalamazoo Assessment Tool

Establish rapport; gather information; understand patient’s perspective; information sharing and consent; closing the session

Items = 20 Score = 100

*Jiang et al. [41] (HMPS) **Wang et al. [48] (Role-playing) ***Chen et al. [32] (SP)

SEGUE

Set the stage; elicit information; deliver information; understand patient’s perspective; end the encounter

*/**Items = 25 */**Score = 25 ***Items = 20 ***Score = 100

Pan et al. [36] (FHM)

Four Habits Coding Scheme (FHCS)

Items = 32 Score = 160

Hang et al. [40] (Role-playing) Yu et al. [30] (SSP)

CS examination

Invest in the beginning; gather information; information feedback; understanding, discussion, and end the session; expression and effects of communication Unclear

Post: Significant improvement on all five domains (p < 0.01); significant improvement on overall qualification rate and score (p < 0.01); Kalamazoo Assessment Tool adapted and translated into Chinese; Likert scale used *Post: Significant CS improvement on all five domains (p < 0.0001) **Post: Significant CS improvement on all five domains (p < 0.05) ***Post: Significant CS improvement on all five domains and overall qualification rate (p < 0.01) SEGUE adapted and translated into Chinese by Li Juan in 2008, reliability and validity had been verified [23]; Likert scale used Post: Significant CS improvement on overall score (T = 741, T > T0.01, p < 0.01) FHCS adapted and translated into Chinese; Likert scale used

CS examination

Communication with patient; information gathering; case analysis; physical check-up; record writing

Score = 100

Wu et al. [42] (HMPS) Zhang et al. [31] (SP)

CS examination

National HMPS Certification Examination Non-verbal; professionalism; empathy; understand and respect patient; emotion handling

Score = 100

Shi et al. [33] (SP&SFM)

Self-designed scale CS examination

Items = 5 Score = 100

He et al. [43] (HMPS)

Self-designed scale

Politeness and empathy; attentive listening; information delivering; cause and cost explaining; shareddecision making Establish rapport; empathy; patient-centered information gathering; information delivering

Unclear

R: Greater improvement on empathy (p = 0.021)

Wang et al. [47] (Role-playing)

Self-designed questionnaire

Unclear

Post: Decreased difficulty level of communication, and improved willingness of application of CS (p < 0.01); improved CS (100%)

Wu et al. [42] (HMPS)

Self-designed questionnaire HMPS Patient– Doctor Location Assessment

Difficulty of communication; application of learnt communication skills; emotional support to patient; patient management Knowledge of HMPS; importance of HMPS training; need of HMPS training; suitable time of HMPS training; HMPS equals doctor–patient communication; HMPS is helpful to improve medical professionalism Doctor’s authority; physical check-up; trust and respect; patient’s culture and background; importance of interacting with patients; patient’s seeking for information; patient’s describing of illness

Items = 5 Items = 18

Post: improved HMPS knowledge, need of HMPS training and suitable time of HMPS training (p < 0.05); No difference on HMPS is helpful to improve medical professionalism and HMPS equals doctor–patient communication (p > 0.05) Post: 10 of 18 items of great difference (p < 0.05), which indicated a change from ‘‘doctor-centered communication’’ to ‘‘patient-centered communication’’ after the training (55.5%)

Trainee’s self-confidence on CS improvement (studies = 11)

Self-designed scale

Score = 100

Unclear

R: Higher CS examination score (u = 3.721, p < 0.05) R: Higher score on communication with patient, information gathering, case analysis, and physical check-up (p < 0.01); no difference on record writing (p > 0.05) Post: Good exam score (mean = 87.5) Post: Both research group and control group obtained CS improvement (p < 0.05) R: Greater improvement on non-verbal, professionalism, empathy, and understand and respect patient (p < 0.001); no difference on emotion handling (p > 0.5) Post: Overall score good (mean = 84.12), highest score on politeness and empathy (=90.95)

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

7

Table 2 (Continued ) Outcome

Study/training model or method

Assessment instrument

Domain measured

Item and score

Comments on results

Lai et al. [44] (HMPS)

HMPS Patient– Doctor Location Assessment Narrative selfreport

Items = 18

Post: Improved CS; better understanding of patient; improved on establishing partnership with patient; more humanistic care for the patients

Li et al. [35] (Roleplaying)

Self-designed questionnaire

Doctor’s authority; physical check-up; trust and respect; patient’s culture and background; importance of interacting with patients; patient’s seeking for information; patient’s describing of illness Prepare for communication; establish rapport; gather information; explain disease; reach consensus; close communication

Items = 30

Sun et al. [34] (PBL)

Self-designed questionnaire

Knowledge commanding; empathy

Unclear

Wang et al. [48] (Role-playing)

Self-designed questionnaire

Unclear

Hang et al. [40] (Role-playing)

Self-designed questionnaire

Wuensch et al. [38] (SPIKES)

Visual Analog Scale (VAS)

Sun et al. [39] (SPIKES)

Breaking bad news Assessment Schedule (BAS)

Establish rapport; gather information; explain treatment plan; inform and provide patient with optional treatments; evaluate patient’s compliance; encourage shared decision-making; acknowledge patient’s endeavors; end the consultation Difficulty of interaction with patient; willingness to improve CS; communication skills improvement Feel secure on breaking bad news; feel secure talking prognosis; feel secure talking prognosis with family; feel secure talking death; feel secure talking death with family; emotional support to patient; emotional distress; quality of consultation; satisfied communication with patient; satisfied communication with family; theoretical knowledge on breaking bad news; applying theoretical knowledge Set the scene; check information patient already knew; patient’s willingness of information obtaining; patient’s competence on accepting and discussing cancer; open-ended question; patient/family’s fear or upset; patient/family’s silence; patient/family’s nonverbal; avoid jargon; patient/ family’s understanding of information; emotional support to patient; empathy

Post: R – Improved on establishing rapport, gathering information, and closing communication (p < 0.05); C – Improved on establishing rapport, gathering information, and explaining disease (p < 0.05) R: Greater improvement on establishing rapport (p < 0.05) No other major difference between the two groups R: Greater improvements on knowledge commanding (p = 0.018) No difference on empathy between the two groups (p > 0.05) Post: Improvement on six domains (p < 0.05) except gathering information and explaining treatment plan (p > 0.05)

Blatt et al. [45] (PCC)

Zhu et al. [46] (Role-playing)

Unclear

Items = 12 (Scale =10 cm)

R: Lower level of difficulty when interacting with patients; greater CS improvement; better information gathering Post: Significant CS improvement on 11 domains (p < 0.05) except emotional distress Cross culture barriers: Western patient autonomy vs. China’s family decision making principle; in 78% of the cases, Chinese doctor informed the family member first Bias of results could not be excluded because of high attrition rate (38%)

Items = 12 Score = 60

Post: Significant CS improvement on 9 items (t = 2.67, p < 0.02), no improvement on open-ended question (p = 0.20), empathy and providing emotional support to patient (p = 0.10) BAS adapted and translated into Chinese (Cronbach’ a = 0.82, CVI = 0.91); Likert scale used

Narrative self-report

–

Narrative self-report

–

Post: Better understand of the importance of doctor–patient communication; improved communication knowledge and skills Cross culture barriers: Western patient autonomy vs. China’s family decision making principle; Chinese doctor’s traditional authority; difficulty on dealing with strong emotion disclosing; Resistance to interactive small group learning Post: Improved CS; decreased discrimination on patient with HIV/AIDS

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

8 Table 2 (Continued ) Outcome

Study/training model or method

Assessment instrument

Domain measured

Item and score

Comments on results

Trainee’s attitudes toward training (studies = 10)

Sun et al. [39] (SPIKES)

Self-designed questionnaire

Unclear

Post: Positive appraisal 61–83%

Wuensch et al. [38] (SPIKES) Yu et al. [30] (SSP)

Self-designed questionnaire Self-designed questionnaire

Knowledge; role-playing; experience sharing; SPIKES model Role-playing; feedback; SPIKES model; workshop SSP

Score = 1–6

Post: Very good

Unclear

Shi et al. [33] (SP&SFM) Sun et al. [34] (PBL) Zhu et al. [46] (Role-playing)

Self-designed questionnaire Self-designed questionnaire Self-designed questionnaire

Post: Positive appraisal 93% A few students (7%) held that SSP were not a good training method, mainly because some SSP did not play the role seriously Post: Positive appraisal 90%

Lai et al. [37] (PTM)

Self-designed questionnaire and Narrative self-report Narrative selfreport Narrative selfreport

Blatt et al. [45] (PCC) Wu et al. [42] (HMPS); Lai et al. [44] (HMPS) Patient satisfaction (N = 2)

SP&SFM

Unclear

PBL teaching model

Items = 7

Help to work and life; reduce discrimination to patient with HIV; achieving training goal Role-playing; case study; practice teaching model

Unclear

Post Positive appraisal: R: 76.6% vs. C: 50% Post: Positive appraisal 94%; 96.6%; 100%

Unclear

Post: Positive appraisal 92–96%

Interactive teaching approach

–

Post: Positive appraisal 65%

HMPS model

–

Post: Like the HMPS training model

Post: High patient satisfaction rate (100%) on understanding disease, obtaining information from doctors, and doctor’s attitude R: Higher patient satisfaction score on interns’ work attitude (p = 0.012) Likert scale used

Pan et al. [36] (FHM)

Self-designed questionnaire

Unclear

Unclear

He et al. [43] (HMPS)

Self-designed questionnaire

Unclear

Score = 10

CST: communication skill training; CS: communication skills; SP: standardized patient; SSP: student standardized patient; SFM: standardized family member; HMPS: humanistic medicine practice skill; FHM: four habits model; PCC: patient-centered communication; PBL: problem based learning; SPIKES: Setting, patient Perception, Information, Knowledge, Empathy, Summary and strategy); PTM: practice teaching model.

contents in the same studies. Six studies [29,32,36,39,41,48] used the Likert scale when developing the instruments. Unfortunately, we could not obtain further information on the modification process of the instruments except the domains and items. 3.3.2. Self-developed scales or questionnaires, examinations and narrative reports Ten [31,33–36,40,42,43,47,48] of the 20 studies used selfdesigned questionnaires or scales as their assessment instruments, and 9 of the 10 did not test the reliability and validity. The CMDA developed a questionnaire called the ‘HMPS Patient–Doctor Location Assessment’ which has 18 items, and was applied in two studies [42,44]. Except for the HMPS questionnaire, no selfdesigned questionnaires or scales were repeated in the 20 included studies. Post-training examination was implemented in three of the studies [30,40,42]. Four studies [37,44–46] applied narrative reports as their measurements. Three studies [38,40,43] used the Likert scale and one [38] used a Visual Analog Scale as the basis for the scales or questionnaires. 3.4. Effects of communication skills training All of the included studies evaluated the effects of CST immediately after the training, except one study [29] which assessed the effects 2 months after the CST was completed; and only one study evaluated the effects of CST in actual practice in clinical settings [36]. Most of the included studies used t-test and/or x2-test for statistical analysis of the data, and some used ANOVA. For those studies that used scales or questionnaires to evaluate the effects of CST, P values were shown to indicate the

significance of any differences (Table 3). The outcomes of CST and the main findings in the studies are categorized in Table 2. 3.4.1. Reported changes in trainees’ communication skills A moderate to significant improvement in communication skills was reported in 11 studies [29–33,36,40–43,48]. Four studies [29,32,41,48] reported significant improvements in five domains including establishing rapport, gathering information, understanding the patient’s perspective, delivering information, and closing the encounter. A meta-analysis was conducted for the three studies that used the SEGUE assessment scale [32,41,48], and the pooled estimate of effect was 10.98 (95% CI: 15.37 to 6.59; P < 0.00001), indicating a significant improvement in communication skills following the training. However, the number of studies included in this calculation was too small for this analysis to be conclusive. Trainees attained significant improvements in communication skills after the Four Habits Model training [36]. The research group that received CST achieved greater improvement in empathy [31,43], non-verbal, professionalism, and understanding patients than the control group [31]. Four other studies reported higher scores after training based on the results of the scale assessing communication skills [30,33,40,42]. Of these 11 studies, 9 [29,32,33,36,40–43,48] employed a combination of training methods including SP, role-playing, feedback, case study, lectures, and group discussions, while only 2 studies used a single training method of SP or SSP [30,31]. 3.4.2. Trainees’ self-confidence in communication skills improvement Despite the majority of studies showing an improvement in self-confidence in communication skills, a few studies reported

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

9

Table 3 Risk of bias for included studies (studies are organized in chronological order within study design). Study

Study design

Participants selection strategy

Grouping strategy

Participation rate

Attrition rate

Intervention measures other than comparable measures

Statistical analysis

Reliability or validity of assessment tool

Yu et al. (2010) [30] Zhang et al. (2010) [31] Hang et al. (2009) [40]

RCT RCT CCT

Convenience Convenience Random

Random Random Convenience

100% 100% 100%

0% 0% 0%

Differenta Same Differentb

Not reported Not reported Not reported

Sun et al. (2011) [34] Li et al. (2011) [35] He et al. (2013) [43] Pan et al. (2009) [36] Lai et al. (2009) [44] Wu et al. (2010) [42] Jian et al. (2010) [29]

CCT CCT CCT PPTD PPTD PPTD PPTD

Convenience Convenience Convenience Convenience Convenience Convenience Convenience

Convenience Convenience Convenience Randome – Conveniencef –

100% Not reported Not reported Not reported Not reported 100% 100%

0% Not reported Not reported Not reported Not reported 4.9% 0%

Differentc Differentd Same Same Same Same Same

t-test t-test ANOVA u-test X2-test t-test t-test t-test Not reported X2-test t-test X2-test Spearman correlation

Wang et al. (2010) [47] Jiang (2011) [41] Sun et al. (2012) [39]

PPTD PPTD PPTD

Convenience Convenience Convenience

– – –

100% 100% 100%

0% 0% 0%

Same Same Same

X2-test t-test t-test

Wang et al. (2012) [48]

PPTD

Convenience

–

Not reported

Not reported

Same

Chen et al. (2013) [32]

PPTD

Convenience

–

100%

0%

Same

t-test X2-test ANOVA t-test X2-test t-test –h –i Not reported Frequency

Wuensch et al. (2013) [38] Blatt et al. (2009) [45] Lai et al. (2010) [37] Shi et al. (2010) [33] Zhu et al. (2011) [46] a b c d e f g h i j

PPTD PTD PTD PTD PTD

Not reported Not reported Not reported Convenience Convenience

– – – – Conveniencej

Not reported 100% 100% 100% 100%

g

38.7% 0% Not reported 0% 0%

Same Same Same Same Same

Not reported Not reported Not reported Not reported Not reported Not reported r = 0.55 and r = 0.63 for pre–post scoring on communication skills and clinical skills; r = 0.62 for improvement synchrony Not reported Not reported Cronbach’s alpha = 0.82, CVI = 0.91 Not reported

Not reported Not reported – – Not reported Not reported

Research group: applied SSP and ESP; control group: applied traditional teaching. Research group: applied CST; control group: no CST. Research group: applied PBL; control group: applied traditional teaching. Research group: applied role-playing; control group: applied traditional teaching. Participants were randomly divided into four sub-groups receiving same CST. Participants were divided into four sub-groups receiving same CST. The authors stated a bias in the results because of high attrition rate. Results were synthesized based on participants’ narrative report. Some narrative reports on improved communication skills after CST. The participants were divided into health care provider group and patient group.

negative findings. Decreased difficulty in interacting with patients [40,47], enhanced communication skills [35,37–40,44–48], better communication knowledge [34,45], and improved doctor–patient relationship [44] were reported. The participants believed the learner-centered training program was helpful in improving their communication skills and clinical competence [47]. A change from ‘doctor-centered’ communication to ‘patient-centered’ communication was demonstrated in two studies [42,44]. Trainees experienced no improvements in demonstrating empathy [34,39], asking open-ended questions, and providing emotional support to patients [39]. A bias in the study findings was reported due to the high attrition rate (38%) [38]. 3.4.3. Trainees’ attitudes toward training Ten studies [30,33,34,37–39,42,44–46] evaluated trainees’ attitudes toward the communication training, and all obtained moderate to high acceptance or satisfaction rates. The participants found the workshop to be very successful [45]. Standardized patient (SP) was highly accepted by the participants [33]; and roleplaying, feedback and group discussions were welcomed by the trainees [38,39,45]. Physicians who received HMPS or SPIKES training said they liked these training modes very much [38,42]. Both the physician and patient groups agreed that the

goal of enhancing communication skills was achieved, and the physicians also stated they experienced decreased stigma when treating patients with HIV/AIDS [46]. However, a few students (7%) found that the student standardized patient (SSP) was not a good training method, mainly because they thought some of the SSPs had not played the role seriously [30]. 3.4.4. Patient satisfaction Only two studies [36,43] evaluated patient satisfaction following the interaction with trainees after CST. A patient satisfaction survey was conducted after the Four Habits Model training, which indicated that patients were 100% satisfied with the physicians on their understanding of the patient’s illness, doctor’s attitude, and degree of information sharing [36]. Interns who received CST obtained a higher score of work attitude from patients than those without CST [43]. 4. Discussion and conclusion 4.1. Discussion We critically evaluated 20 studies focusing on communication skills training (CST) for physicians and medical students in

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 10

X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

mainland China. The results demonstrated that the majority (90%) of the CST program improved trainees’ communication skills by applying a training design involving a didactic component (lecture) combined with practical rehearsal (SP, role-playing, scenario simulation) and feedback (group discussion). Based on the reported results of these 20 studies, the most effective components of CST which improved the communication skills of Chinese doctors and medical students were role-playing and SPs. The mean duration of short-term CST was approximately 17 sessions over 2.5–3 days. The training strategy was consistent across the studies when measuring this aspect of CST [55–57]. Four of the 24 assessment instruments were adapted from Western countries, and 20 of the 24 instruments were self-developed in which 19 were used in only one study each for a decade and have never been validated. The low rates of validation reflect the results attained 20 years ago [58]. With regard to trainees’ attitudes toward communication training, moderate to significant positive appraisal was obtained. This indicates a relatively high willingness to accept CST and its effects on improving communication skills among trainees. The main goals of the doctor–patient CST are to enhance the doctor’s competence in communication skills, to improve patient satisfaction and health outcomes, and to facilitate the relationship between patient and doctor. Based on the included studies, the outcomes of CST were characterized by significant enhanced competence in communication skills and self-confidence in the trainees. Participants showed more respect and demonstrated better understanding toward patients after receiving CST. In addition, their relationship with patients improved. However, negative findings such as no improvement in open-ended questions and empathy were also reported [34,39]. These may be due to the influence of doctors’ traditional interviewing habits and the fact that physicians tend to keep their emotional expression hidden from patients [39]. The study [30] which applied the student standardized patient (SSP) was also questioned by a few trainees who thought that the use of the SSP was not a very good training approach. Thus, we think future CST in China should place more emphasis on the development of training methods. The training contents and strategies were different according to the training objectives. For example, the content on responding to emotions with empathy and the strategy of involving patient’s family members were added in studies that focused on breaking bad news. The evaluation of CST effectiveness in the included studies tended to focus on assessing the cognitive aspects of communication. Socio-emotional aspects of communication such as nonverbal were rarely assessed and lacked the application of nonverbal assessment instruments. Few studies evaluated the long-term effects of CST or conducted the assessment in real clinical encounters. It was difficult to reach a conclusion on the relationship between the training model and the improvement in communication skills across the 20 studies. For example, 11 studies tested improvement in trainees’ self-confidence in communication skills. However, each study used a different method for measuring the level of self-confidence and these methods were too dissimilar to combine across studies. This indicated that China is still at the first-stage of research on CST, and more in-depth studies are needed. The studies included in this review illustrated the potential relationships between patient satisfaction and CST, and an improved satisfaction rate was reported. However, only two studies evaluated patients’ perception of change in the trainees’ attitude attributable to CST. None of the included studies evaluated patient health outcomes such as adherence and symptom resolution. As patient satisfaction and health outcomes are the key indicators for improving doctor–patient rapport, we

recommend that future CST studies in China should include these evaluation indices. Furthermore, an important goal of a communication training model is how to respect a patient’s rights (such as informed consent). Good communication skills in a physician should be evaluated by providing correct and full treatment information for patients, improving the quality of medical service and the trust between doctor and patient and deceasing medical errors. Future studies on CST should place more emphasis on developing training models that can meet these requirements. Communication skills training has only been available in China for approximately 10 years, and many training models and assessment instruments of CST were originally borrowed from Western countries as demonstrated by the included studies. However, even though the Calgary–Cambridge guideline [63] and Liverpool Communication Skills Assessment Scale (LCSAS) [64] have been widely known in the field of communication training in China [62,65], they were not used in the included papers. Furthermore, cultural differences exist between China and Western countries. Cross-cultural barriers such as Chinese family decision-making tradition vs. Western individual autonomy, and Chinese doctor’s manner of traditional authoritative manner vs. Western doctor’s patient-centered communication have been reported by some studies [42,45]. These findings were echoed in the study by Fritz et al. who demonstrated a hierarchical doctor– patient relationship in Asian countries including China. Difficulties with, and challenges in, accommodating the CST intervention to the Chinese medical environment are inevitable. Researchers are advised to focus more on adaptation of the Western CST model and assessment instruments to narrow the cross-cultural gap. Five major limitations of this review should be noted. First, we did not include unpublished or gray literature in our review, and this may have introduced some publication bias. However, based on an unsystematic review of conference reports relating to CST, we think our conclusions from this review are unlikely to have been affected by excluding unpublished research, as most of the results in these publications were similar to the included studies. Second, selection bias may have occurred as only medical students and practice physicians were included as target participants in our review. Although we had initially intended to include nurses and other medical professionals in the study population, we discovered that CST for nurses is quite different to that for doctors and medical students. Third, problems associated with heterogeneity of the sample may have occurred due to differences in the medical education system in China and that in Western countries, especially in education tracks design and the specialty of curriculum allocation. Two tracks have been developed in China, one is ‘‘Five-year Medical Education’’, and the other is ‘‘Eight-year Medical Education’’. ‘‘Five-year’’ medical students graduate with a bachelor degree, and they can either become a practice doctor or pursue a MD after graduation. ‘‘Eight-year’’ medical students graduate with a MD. However, as the ‘‘Eight-year’’ track is actually an extension of the ‘‘Five-year’’, students from both tracks can be called ‘‘undergraduates’’. Furthermore, curricula of the specialty of ‘‘clinical medicine’’ and ‘‘clinical psychology’’ are quite similar in China; hence, future psychologists probably receive the same doctor–patient CST as other medical students. The situation is quite different in other countries. For example, in the U.S., students receive undergraduate education in specialties other than medicine, and then seek a MD. Fourth, as a number of the included studies did not provide enough information or differed from each other extensively, we were unable to compare the effects of CST across these studies with regard to certain characteristics such as training models, training duration, instructional methods, and assessment tools. Fifth, even though the included studies were of relatively good scientific quality, and the basis and results of this systematic review are reliable, the scientific quality of some of the

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

included Chinese articles was not as good as the English articles in the same field. This may be because the research into communication in health care has a relatively short history in China. However, we were encouraged to see that increasing numbers of Chinese researchers and educators are devoting their efforts to doctor–patient communication training, and the quality of their studies has improved over the years. We are quite confident that in the near future there will be higher quality Chinese articles on communication skills. 4.2. Conclusion Our systematic review is the first to synthesize, compare, and analyze published research on CST for physicians and medical students in mainland China. The variety of studies in this review reflects the various communication skills training strategies, the wide range of assessment instruments, and the outcomes that CST may influence. Although our findings offer an evidence-based demonstration that CST can enhance patient–physician communication in China, additional research is needed to better understand the effects of CST on the physician–patient relationship, patient outcomes and the contributions of specific training methods to the desired training objectives. The results of this review and our recommendations for future studies provide an important starting point for this work in mainland China. 4.3. Implications for research Competence in doctor–patient communication is a key skill in clinical physicians. The basic ethical principles of communication between doctor and patient include trust and respect, equity, confidentiality, and informed consent [59]. Given the frustrating patient–doctor conflicts in mainland China, an enhanced patient– medical professional relationship should be a public health priority that requires effective training of medical students and physicians in a responsive, informed and empathetic doctor–patient interaction. To improve the physician’s and medical student’s communication skills, strictly planned, culturally competent, effectively implemented and rigorously evaluated training is required. Based on our review, we suggest that future studies of CST in mainland China will improve research quality and relevance for practice by following these guidelines: (1) employ a training model with sound evidence of effectiveness; (2) use multiple training methods including a combination of lectures, practice rehearsal and feedback; (3) apply assessment instruments with demonstrated reliability and validity; (4) evaluate a trainee’s competence in communication skills as well as patient outcome including patient satisfaction, adherence, and symptom resolution; (5) assess both the short-term and long-term effects of CST, including observation of actual clinical interactions between the patients and CST-trained doctors; (6) include specific Chinese cultural elements into the CST training programs to enhance acceptance by clinical staff and administrators. We recommend that Chinese medical universities pay more attention to doctor–patient communication training for medical students. Finally, we encourage administrators in Chinese hospitals to incorporate doctor–patient communication skills training as a key and required component of continuing education for practicing physicians. Funding This study was funded by the New York Chinese Medical Board G16917342 (CMB). The founders had no role in the design and conduct of the study, the collection, management, analysis and interpretation of data, or the preparation, review or approval of the manuscript.

11

Conflicts of interest None. Ethical approval Not required. Acknowledgements The authors wish to thank Adriaan Visser, an expert on communication skill training of EACH, for his contribution to the critical review of this paper; the authors wish to thank Marc Van Nuland, a communication skills lecturer at the University of Leuven, for his contribution to the proofreading of the paper; the authors also wish to thank Professor Gerry Humphris of the University of St. Andrews Medical School, for his contribution to the final review of the paper. References [1] Kurtz S, Silverman J, Draper J. Teaching and learning communication skills in medicine. 2nd ed. London, UK: Radcliffe Publishing Ltd.; 2009. [2] Stewart MA. What is a successful doctor–patient interview? A study of interactions and outcomes. Soc Sci Med 1984;9:167–75. [3] Kaplan SH, Greenfield S, Ware JE. Assessing the effects of physician–patient interactions on the outcomes of chronic disease. Med Care 1989;27:S110–27. [4] Levinson W, Lesser CS, Epstein RM. Developing physician communication skills for patient-centered care. Health Affair (Millwood) 2010;29:1310–8. [5] Griffin SJ, Kinmonth AL, Veltman MWM, Gillard S, Grant J, Stewart M. Effect on health-related outcomes of interventions to alter the interaction between patients and practitioners: a systematic review of trials. Ann Fam Med 2004;2:595–608. [6] Levinson W, Roter DL, Mullooly JP, Dull V, Frankel R. Physician–patient communication: the relationship with malpractice claims among primary care physicians and surgeons. J Amer Med Assoc 1997;277:553–9. [7] Levinson W. Physician–patient communication: a key to malpractice prevention. [Editorial]. J Amer Med Assoc 1994;272:1619–20. [8] Beckman HB, Markakis KM, Suchman AL, Frankel RM. The doctor–patient relationship and malpractice. Lessons from plaintiff depositions. Arch Intern Med 1994;154:1365–70. [9] Wofford MM, Wofford JL, Bothra J, Kendrick SB, Smith A, Lichstein PR. Patient complaints about physician behaviors: a qualitative study. Acad Med 2004;79: 134–8. [10] Silverman J, Kurtz S, Draper J. Skills for communicating with patients. 2nd ed. London, UK: Radcliffe Publishing Ltd.; 2008. [11] Rotthoff T, Raehring T, David D, Bartnick C, Linde F, Willers R, et al. The value of training in communication skills for continuing medical education. Patient Educ Couns 2011;84:170–5. [12] Aspegren K. BEME Guide No 2: teaching and learning communication skills in medicine – a review with quality grading of articles. Med Teach 1999;21: 563–70. [13] Hulsman RL, Ros WJ, Winnubst JA, Bensing JM. Teaching clinically experienced physicians communication skills. A review of evaluation studies. Med Educ 1999;33:655–68. [14] Maguire P. Can communication skills be taught? Br J Hosp Med 1990;43:215–6. [15] Li S, Liu Y. The achievements, problems and experiences of the health service development in China’s 30 year reform and opening-up. Chin J Health Policy 2008;11:3–8. [16] Zheng XQ, Deng LQ, Chen CL. An investigation on the medical dissensions and infringement events in 326 medical institutions. China Hosp 2002;6:24–30. [17] Chinese Doctor Association. An investigation on ‘medical riots’ among some provinces. City China Health Industry; 2007. [18] The Lancet. Chinese doctors are under threat. [Editorial]. Lancet 2010;376:657. [19] Li B, Sun XY, Wang JF. A review of doctor–patient communication barriers research. China Health Serv Manage 2009;5:303–4. [20] Wang JF, Yin M. Doctor–patient communication. 3rd ed. China: People’s Medical Publishing House; 2013. [21] Peng L, Ran SJ. Teaching doctor–patient communication for medical students: a situation and reflection on curriculum design. Chongqing Med 2011;40:2594–5. [22] Ceng GC, Hong DW. Doctor–patient communication education in hospital. Zhejiang Med Educ 2007;6:21–3. [23] Juan L. Using the SEGUE framework to assess Chinese medical students’ communication skills in history-taking. China Medical University. CNKI dissertation database [dissertation, accessed 22.05.08]. [24] Wang JF. Reconstruction of the doctor–patient relationship under China’s market economy. Chin HospMang J 2006;8:5–8. [25] Beverley JS, Jeremy MG, George AW, Maarten B, Neil A, Candyce H, et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Metho J. Available from: http:// www.biomedcentral.com/1471-2288/7/10 [accessed 10.07.07].

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012

G Model

PEC-4894; No. of Pages 12 12

X. Liu et al. / Patient Education and Counseling xxx (2014) xxx–xxx

[26] Lisa B, Drummond R. The Cochrane Collaboration: preparing, maintaining, and disseminating systematic reviews of the effects of health care. J Amer Med Assoc 1995;274:1935–8. [27] Qu Y, He L, Liu M. Basic review of methods for Cochrane systematic review. Chin J ClinReh J 2003;7:532–6. [28] Fritzsche K, Scheib P, Ko N, Wirsching M, Kuhnert A, Hick J, et al. ASIA-LINK Workgroup results of a psychosomatic training program in China, Vietnam and Laos: successful cross-cultural transfer of a postgraduate training program for medical doctors. BioPsychoSocial Med 2012;6:1–14. [29] Jiang SF, Zhang Y, Zhou R, Zhu SZ. Application of standardized patients in training of communication skills in consultation for resident general practitioners. Chin J Gen Pract 2010;9:525–8. [30] Yu CY, Chen Y, Yang P, Jiang ZQ, Yu ML, LI X, et al. A combination of undergraduate medical student standardized patients with electronically standardized patients and its application to medical students pre-internship training. China Higher Med Educ 2010;10:78–80. [31] Zhang AQ, Chen H, Li L, Wang LG, Ji HP, Xue CY. Evaluation of the validity of standardized patients on clinical communication skills training. Res Med Educ 2010;97:89–91. [32] Chen S, Shi HY, Chen J, Huang H, Ma YM, Li R. Physician–patient communication skill training for community general practitioners. Chin Commun Doc 2013;15:385–6. [33] Shi P, Ren B, Shi JH, Liu YX. Training and assessment of doctors–patient communication skills: establishment of an efficient mode. Med Philos: Hum Soc Med Edi 2010;31:26–7. [34] Sun T, Gao Y, Cheng YT, Li ZZ, Zhang JM. Comparative study of the problembased learning teaching method and traditional teaching model to explain the disease in the doctor–patient communication. China Med 2012;7:1176–8. [35] Li Y, Wang JF, Xu NZ, He DJ, Wu YQ, Wang HF. Application of role playing teaching strategy in developing doctor–patient communication skills of medical students. Chin J Gen Pract 2011;9:1809–10. [36] Pan YH, Ao Y, Tang JZ, Zhao SQ, Liu WW. Investigation of the training target system in communication between ophthalmologist and patient. IMHGM 2009;15:8–10. [37] Lai Y, Chen DB, Dai LX, Zeng J. Strengthening the practice teaching to improve medical students’ doctor–patient communication. Res Med Educ 2010;9: 1113–5. [38] Wuensch A, Tang LL, Goelz T, Zhang YN, Stubenrauch S, Song LL, et al. Breaking bad news in China – the dilemma of patients’ autonomy and traditional norms. A first communication skills training for Chinese oncologists and caretakers. Psycho-Oncol 2013;22:1192–5. http://dx.doi.org/10.1002/pon.3112. [39] Sun YQ, Sun BF, Cui Y, Wen YN, Hao J, Zhao J, et al. An enlightenment from cancer communication skills training workshops to doctor–patient communication curriculum in medical college. Med Philos 2012;33:77–9. [40] Han ZY, Ouzhu LB, Zhang L, Xin YN. Enhancing doctor–patient communication skills training among medical interns in Tibet: a practice and discussion. HlthVocEduc 2009;27:96–8. [41] Jiang Y, Chen JG, Huang YJ, Huang Z. Evaluation on the effectiveness of general practitioners’ social psychological professional skills training. Prev Med Trib 2011;17:399–401. [42] Wu XH, Dong XC, Hu XW, Jiang Y. A research on humanistic medicine practice skill training for residents and specialists. Chin HospMgmt 2010;30:33–5. [43] He W, Shao Y, Zhao JP, Shen ZJ. The significance of doctor–patient communication training based on humanistic medicine for interns in department of urology. China High Med Educ 2013;42–4.

[44] Lai YH, Huang QH, Li LN, Li YM. Humanistic medicine practice skills training for medical doctor. Med Philos: Hum Soc Med 2009;30:13–6. [45] Blatt B, Kallenberg G, Lang F, Mahoney P, Patterson J, Dugan B, et al. Found in translation: exporting patient-centered communication and small group teaching skills to China. Med Educ Online 2009;14:1–8. [46] Zhu XZ, Yao HJ, Gu KK, Wang H, Gao J, Xue YL, et al. Doctor–patient communication barriers in AIDS prevention and treatment and the effect of intervention programs, Shanghai. J Environ Occup Med 2011;28:656–9. [47] Wang C, Wang L, Zhou JF, Cao BL, Zhang HF, Wang HC. The observational study of the professional training effect on doctors of Tianlin community, Xuhui District, Shanghai. Chin Gen Pract 2010;131732–734. [48] Wang Y, Tang Y, He Y, Zhu YQ. Effect of doctor–patient communication education on oral clinical practice. Shanghai J Stomatol 2012;21:474–7. [49] Fetzer B. Conference on physician–patient communication in medical encounters. Essential elements of communication on medical encounters: the Kalamazoo consensus statement. Acad Med 2001;76:390–3. [50] Makoul G. The SEGUE framework for teaching and assessing communication skills. Patient Educ Couns 2001;45:23–4. [51] Frankel RM, Stein T. Getting the most of the clinical encounter: the four habits model. Perm J 1999;3:79–88. [52] Baile WF, Buckman R, Lenzi R, Glober G, Beale EA, Kudelka AP. SPIKES – a sixstep protocol for delivering bad news: application to the patient with cancer. Oncologist 2000;5:302–11. [53] Krupat E, Frankel R, Stein T, Irish J. The four habits coding scheme: validation of an instrument to assess clinicians’ communication behavior. Patient Educ Couns 2006;62:38–45. [54] Miller SJ, Hope T, Talbot DC. The development of a structured rating schedule (the BAS) to assess skills in breaking bad news. Br J Cancer 1999;80:792–800. [55] Berkhof MH, Rijssen JV, Schellart AJM, Anema JR, Beek AJVD. Effective training strategies for teaching communication skills to physicians: an overview of systematic reviews. Patient Educ Couns 2011;84:152–62. [56] Gysels M, Richardson A, Higginson IJ. Communication training for health professionals who care for patients with cancer: a systematic review of effectiveness. Support Care Cancer 2004;12:692–700. [57] Lian Y, Wang YJ, Sara L, Zhang SR. A study of components associated with effectiveness of medical staff–patient communication education program. Chongqing Med J 2009;38:2963–7. [58] Boon H, Stewart M. Patient–physician communication assessment instruments: 1986 to 1996 in review. Patient Educ Couns 1998;35:161–76. [59] Tian F. Ethical reflection on physician–patient communication. Chin Med Eth J 2009;22:106–28. [60] Zhang XQ, Margaret SF. Tensions between medical professionals and patients in mainland China. Camb Quart Hlthc Eth J 2011;20:458–65. [61] Wang LQ. Discussion on continuing education of doctor–patient communication for hospital health care providers. Med Edu Pro J 2011;32:60–1. [62] Xu T, Dong EH, Liu W, Liang Y, Bao Y. Reliability and validity of the Chinese version of the Liverpool Communication Skills Assessment Scale. Chin Psych J 2013;27:829–33. [63] Kurtz SM, Silverman JN, Draper J. Teaching and learning communication skills in medicine. Oxford, UK: Radcliffe Medical Press Ltd.; 1998. [64] Humphris GM, Kaney S. Assessing the development of communication skills in undergraduate medical students. Med Educ 2000;35:225–31. [65] Wu X, Chen JC, Liu JM. Application of Calgary–Cambridge guide in doctor– patient communication skills teaching. China Med Educ Technol J 2013;2: 218–20.

Please cite this article in press as: Liu X, et al. Doctor–patient communication skills training in mainland China: A systematic review of the literature. Patient Educ Couns (2014), http://dx.doi.org/10.1016/j.pec.2014.09.012