The use of a rating instrument to teach and assess communication skills of health-care workers in a clinic in the Western Cape

Research in health communication shows communication to be an important as­ pect of successful health-care. Moreover, training courses which provide feedback have been shown to improve health professionals’ ability to conduct successful interviews. This article describes a rating instrument which was developed in order to facilitate teaching and assessing the communication aspects of health-care inter­ views. The instrument was found to be useful in a training programme offered to nursing staff of a TB Clinic in Mitchells Plain, Western Cape. The instrument ap­ pears as Table 1. In the Table categories of communication behaviours, each indi­ cating an important aspect of the interaction, are given as the six headings. These are: • establishing rapport and respect • listening receptively • confirming the patient • sharing control • informing effectively and checking perceptions.


Research in health communication has
shown the crucial role that communication plays in the successful treatment of a patient.
Within each category the more detailed specific behaviours are listed, allowing for close analysis of a care-giver's interviewing skill.The article briefly discusses the importance in effective communication between the care-giver and patient of each category of behaviours given in the instrument, supported by evidence from re search.Lastly the article describes a "case study" on how the instrument has been successfully used in a training programme.

Introduction
Research in health communication has shown the crucial role that communica tion plays in the successful treatment of a patient (Ley 1982:241-254, DiMatteo et al 198 6:5 8 1-59 4, K orsch 1989:5-9, B o rc h e rd s 1987:5-9, T ho m p so n 1984:146-163).Patients lack the tech nical expertise to judge the quality of a health-care encounter from a diagnos tic or therapeutic point of view.Their satisfaction with the encounter is based on their perception of and response to the quality of comm unication with the h e a lth -c a re p ro v id e r (S tree t et al 1988(S tree t et al :420-440, Street 1989:137-154):137-154).Moreover, such essential factors as pa tients' understanding, recall and adher ence to therapy have all been linked to the communication style of the healthp ro v id e r (Ley 198 2:2 4 1-25 4, Bush 1985:103-117).By contrast, ineffective interpersonal communication has been shown to lead to dissatisfaction with health-care services (Lane 1983:772-799), and can result in alienation be tween health-care professionals and their patients (Korsch 1972:66-74).
In spite of the clear benefit of effective com m unication in successful health care interactions (Kreps 1988:344-359), the tendency among health-care profes sionals is still to interact with the patient in an authoritarian, disease-oriented manner, which leaves the patient's com m unication needs unsatisfied (Wyatt 1991:157-174), and comparatively little attention is paid to this aspect of their training (Korsch 1989:5-9).
There is much evidence that training courses can and do improve communi cation skills (Maguire etal 1978:695-704, Maguire et al 1986:1573-1578, Maguire 1990:215-216).Another finding is that students tend to be poor at judging their own ability to communicate, and need detailed explanation, feedback and prac tice regarding specific communication behaviours (Marteau et al 1991:127-134).This also makes training demand ing and time-consuming.This article describes the rating system, or instrument, designed by the authors to facilitate and evaluate a training pro gramme on the communication aspects of health-care interviews.This empha sis on the interpersonal communication between care-giver and patient is in line with the shift which is prevalent in much contemporary thinking towards a more holistic approach to medical care (Wyatt 1991:157-174, Reiser e ta l 1980, Livesey 1986).The objective of the training was to encourage the nurses to becom e more self-reflective by enabling them to learn and incorporate the complex range of communication behaviours that pro duce a positive interaction with patients.The rating instrument identifies specific behaviours which can make an interac tion more patient-centred, and therefore more successful.The behaviours have been drawn from an analysis of research into the subject of communication be tween care-givers and patients, as well as from the authors' extensive personal experience in teaching communication skills to medical students, nurses and paramedics.Although the instrument still needs to be validated, especially for cross-cultural application, the study shows that it fa cilitated training and that the participants were assessed with a good degree of reliability.The instrument can be used both to pro vide feedback to students in a teaching situation, and for assessment of health care interactions in an exam ination.Teaching personnel often have had no formal training in communication skills themselves and may find the instrument provides useful criteria to apply when teaching and testing students.The instrument differs from previous rat ing scales, such as the scale formulated by Maguire (Maguire et al 1978:695-704) in that it can be used to evaluate a wide range of health-care interviews, not only diagnostic or history-taking interviews.The focus is specifically on communica tion behaviours which can be taught.This emphasis is partly achieved through the six headings, which are deliberately worded as communication activities.In this way both the function and the de sired consequences of the behaviours are immediately apparent.The way in which the behaviours have been organ ized also makes it easy for students to remember and use as a quick mental check while conducting an interview.

The Development of the Instrument Motivation
The instrument was developed to meet a perceived lack in the literature on in terviewing skills in Health Communica tion studies.It has been devised to cover most of the relational aspects in the com munication interaction, while also includ ing the transfer of information.It appears as Table I.

Literature Research
A broad survey of the research dealing with communication aspects of health care interactions was conducted and a comprehensive list of behaviours that had been shown to be characteristic of good patient-centred interviews was col lated and categorised into six clearly defined and useful groupings.These were developed into a rating instrument for both trainers and nurses to use in this study.The intention was to be as com prehensive and consistent as possible in teaching and assessm ent, before, during and after training.The specific behaviours are rated on a 5-point scale as either positive (or "helping" the inter view) or negative ("hindering" the inter view).

Description of the Instrument
The instrument consists of six catego ries of behaviours, which meet different needs of the patient.Each category con sists of specific actions which are con sidered to either help or hinder the inter view.These actions are rated positively (helping the interview) or negatively (hin dering the interview) on a 5-point scale.The six categories of com m unication behaviours that have been identified by the authors are essential for a success ful interview are: 1.
Establishing rapport and respect 2.
Confirming the patient 4.
Checking perceptions In the sections that follow, the behaviours in each category of the instrument are described briefly.

Establishing Rapport and Respect
Establishing rapport with the patient de pends largely on the care-giver's ability to display non-verbal behaviours which extend recognition to the patient.These non-verbal signals are the dominant part of the message which the patient re ceives from the care-giver.Patients are known to be highly sensitive to these non-verbal signals (Bush 1985:103-117).
It is essential to establish a supportive communication climate from the begin ning of the interview, in order to enable the patient to speak openly and to over come tension and shyness.The care giver should try to maintain a posture w hich indicates a relaxed, but alert, frame of mind.Through the expressive use of face, voice and gesture, an atti tude of responsiveness, concern and involvement can be communicated.This expressiveness on the part of the care giver has been identified as one of the major needs in patients ( 0 'Hairetal:125-129).Appropriate eye contact is impor tant in establishing a sense of relation ship, and every attempt should be made to avoid being distracted by medical Curationis June 1999 33 forms and checklists.

Listening Receptively
Listening is a key element in the com munication process (Wolff et al 1983, Van der Merwe 1991) (Chubon 1989:23-38), but should ask questions to clarify the picture.Closed questions such as, " Do you feel better now?" are not likely to elicit a genuine response.Open questions such as " How do you feel now?" can give more insight into the patient's fears, problems and m iscon ceptions (Pendleton et al 1984:8).

Confirming the Patient
Having listened well to the patient, and attempted to clarify the patient's position, the care-giver needs to respond in a way that makes the patient feel accepted and valued.
A confirming response is one that rec ognises the validity of the patient's real ity, and does not place a value judge ment on this reality (Garvin et al 1986:1-19, Steyn, 1994) (Garvin et al 1986:1-19).
To be able to respond in a confirming way requires that the care-giver should not be self-involved; and is a reflection of the care-giver's self-esteem.

Sharing Control
Most patients respond well to interac tions in which there is a sense of equal ity and in which they share the responsi bility for directing the interview (Burgoon et al 1987:307-324, Street et al 1988:420-440, Korsch 1989:5-9).O'Hair (1989:97-115) points out that patients are no longer passive participants in the health care relationship, but "are willing to as sert themselves with messages that at tempt control or neutralize control of the transaction" .He sees the resurgence of malpractice litigation and treatment noncompliance as consequences of this de sire for more com m unicative control.The extent to which patients are com fortable with increased responsibility has been linked to whether the patient has an internal, or external, health locus of control.Patients with an internal health locus of control are most comfortable with increased responsibility (Arntson et al 1989:75-95).To ensure that the real issues at stake for the patient are aired (Holt 1990:131-132), the patient should do more speaking than the doctor or nurse.Any problems can then be met with appropriate information.This does not mean that the interview should lack direction; the care-giver can use prob ing questions to keep the interview on track, and skilfully redirect the patient if there is too much digression.This redi recting should be done by openly tell ing the patient that this is what one is doing.
Whenever possible, the patient should be allowed to set his or her own goals to solve the problems that arise from the illness, thus taking a measure of respon sibility for the cure.This requires that the care-giver should approach issues in the spirit of negotiation and partner ship.Therapy thus negotiated with the patient is more likely to succeed (Korsch 1989:5-9).

Informing Effectively
Patients' satisfaction with medical treat ment has been linked to their under standing of the information presented to them (Ley 1982:241-254).To be effec tively received, information should be presented at the appropriate moment of the interview.The early stages of the interview should be used to release ten sion, and to establish a supportive at mosphere.An "agenda" for the interview should be negotiated with the patient.If this early part of the interview is done ef fectively, the objectives of the middle sec tion of the interview can be better met, when the patient feels more at ease to talk about things that affect him or her deeply.When the patient feels able to talk openly the likelihood of identifying the real issues increases.
The inform ation given to the patient should be appropriately pitched for the patient's level of understanding; unnec essary medical jargon, which is at best confusing, and at worst alarming, should be avoided.Moreover, since patients feel a great need for information (Kreps 1988:344-359), the doctor should not only inform the patient about the disease and the treatment, but also involve the patient in the problem-solving required to make a diagnosis.The information should be introduced at a pace that is appropriate to the patient.
The end of the interview should consist of summarising important information, and formulating an action-plan based on the point of view of both the care-giver and the patient.The responsibilities each has for the success of the treatment should be spelt out clearly.

Checking Perceptions
An analysis of care-giver-patient rheto ric shows that although both care-giver and patient may be addressing the same topics, they may miss each other com pletely (Sharf 1990:217-229).
The process of checking perceptions is most important to the success of the in terview.Health care professionals tend to be unaware of their patients' percep tions of health interactions, and yet it is the patient's perception that influences the outcome of the treatment.Health care professional's self-perception of their com m un icatio n style has been shown to be irrelevant to the patient's satisfaction with the interaction (Street et al 1988:420-440).
A patient is more comfortable if the care giver is perceived as being similar to him or her (Burgoon etal 1987:307-324).The care-giver should therefore try to work with the patient's perceptions, clarifying how the patient understands significant issues and using that understanding as a framework for explanation to the pa tient.This checking of perceptions should occur frequently throughout the interview.
At the end of the interview the care-giver can give a summary of his or her per ceptions of significant moments in the interview, such as when the patient ap peared to be nervous, and ask for feed back on whether the issues are now re solved for the patient.Checking the interview in this way not only ensures congruence of perceptions between care-giver and patient, but also allows for the development of rapport between care-giver and patient.Moreo ver, the care-giver gets an idea of his or her ability to conduct a patient-centred interview.This ensures that the care giver obtains continual feedback on which future personal growth can be based.

Case Study:
A communication train ing programme using the instrument The instrument was first used in a study on the quality of com m unication be tween nursing sisters and tuberculosis patients, at a TB day clinic in Mitchells Plain in the Western Cape (Steyn et al 1997:53).Five nursing sisters on the staff of the Clinic participated.They were not selected for any skills or history, but were the staff routinely allocated for duty on the day the study commenced.
In the study interactions between the nursing sisters and their patients were analysed using the instrum ent.Two video-recordings were made of each sis ter interviewing a newly-diagnosed TB patient; one made before and one after training in communication skills.The 20hour training programme was given by the authors, all of whom were on the staff of the Professional Communication Unit at the University of Cape Town.The rat ing instrument was used in order to as sess th e e ffe ct o f the tra in in g p ro gramme, and was done by six asses sors.Two of the assessors were not di rectly involved in the project and did not know which of the video recordings had been made before and which after the training phase.

Procedure Before and After Training
The patients' permission was requested for the entire interview to be recorded on video.Interviews were conducted in an office in which the video cameras had been positioned so as to carefully ob serve both the nurse and the patient's face and body throughout the interview.
Immediately after the interview a ques tionnaire was administered to the pa tients to obtain their impression of the interview.The nurses were also asked to rate themselves by means of a similar questionnaire.The intention was to pro vide nurses with feedback and opportu nity for reflection on their own skill.

The IVaining Programme
The training p rogram m e began with theoretical input on the principles of ef fective interpersonal com m unication, and the overall functional structure of an interview.The list of behaviours de scribed in the rating instrum ent were explained in detail, and the reason for their significance.The Convergence M odel of C om m unication (Rogers & Kincaid) was found to be useful for illus trating the on-going, dynamic nature of the interview and the desired progres sion toward mutual understanding and equality.

Role Plays
Part of the training was done through role-plays.The following scenarios were used which deliberately focused simply on the emotional and social implications of the disease.Each nurse had the op portunity of taking the part of the nurse, the patient and the observer.

Examples of Role-play Scenarios
Patient's Information

Nurses Information
The patient in front of you is Anne, a 19 year-old unmarried mother with a daugh ter of two.

Patient's Information
You are Mary, a 40 year-old char, work ing Mondays, Tuesdays and Fridays for different employers.You are divorced and have no other support.You live alone with three dependent school go ing children.You cannot possibly come to the clinic on these days for your medi cation as you would risk losing your jobs.You are not going to let your employers find out that you have TB.

Nurses Information
The patient in front of you is Mary, a 40 year-old mother of three.

Use of Video in IVaining
The ro le -p la y s w ere re c o rd e d on videotape and discussed by the nurses and the trainers.The nurses were also shown their first videotaped interviews with patients in the Mitchell's Plain Clinic, so that they could observe their own behaviour as well as the responses of their patients.They used the rating instrument to evaluate their own consultation skills and identify areas needing improvement.Discussion and peer evaluation was encouraged to give them as much helpful feedback as possible.

Evaluation of the Patient Interviews Recorded Before and After IVaining
The two sets of videorecorded patient interviews were first transcribed from the tapes.In order to evaluate the nurses' performance as objectively as possible, six assessors were asked to rate the c o m m u n ic a tio n b e h a v io u rs o f the nurses.Two assessors did the rating 'blind' in that they were not told whether they were watching the 'before or after' video recording.These assessors were communication teachers who were not otherwise involved in the study.
All the assessors used the rating instru ment to assess the six categories of be haviour as either positive (helping the interview) or negative (hindering the in terview) on a 5-point scale.In all a total of 1560 observations were made, a total of 26 for each nurse.The scores ob tained are described in an earlier publi cation in Curationis (Steyn et al 1997 p. 54).The rating instrument was able to be used by all the assessors to judge the c o m m un icatio n behaviours w ith enough understanding and consistency.
Figure 1 shows that all the nurses im proved during the training period.The parallel profiles exhibited by the six as sessors indicate that they were in good agreement.However some differences did occur in the rating.For example, the two 'blinded' assessors involved in the study a p peared more stringent in their evaluation of the nurses performance.Neverthe less, the scores dem onstrate that the raters were consistent in fin ding the same trend to overall improvement; and in identifying the behaviours that were responsible for the trend.

Conclusions
1.Although research is needed to fur ther validate the psychometric proper ties of this instrument, the use of six in dependent assessors does establish in ter-rater reliability for the instrument.2. The instrument was able to be used successfully by two " blind" assessors, which indicates that they were able to understand the terms used to describe the communication behaviours, and also able to identify and interpret these be haviours.
3. The degree of correlation between the scores of the six assessors in the case study indicates that they had been able to use the instrument with a degree of consistency and reliability.Although the "blinded" assessors appeared more stringent in their evaluation of the nurses, all the assessors were in good agree ment using the rating instrument.(Steyn et al 1997).
4. No cross-cultural validation of the in strument has been done, and the instru ment is as yet entirely informed by west ern research on the subject.The instru ment should be used keeping this in mind.Further refinement and validation is now necessary.
5. Clearly the nature of communication in health-care interviews is such that it cannot be satisfactorily described by a mere check-list of behaviours.However, for the purpose of teaching, it is neces sary to be able to convert this daunting complexity into a form that can be un derstood and applied.The instrument was found to be a useful tool for this purpose by both trainers and nurses at all the stages of this study.
6.The usefulness of the instrument is partly due to the way the material is or ganised.The six main categories each encompass a complex set of competen cies and abilities and thus avoids being just a mechanistic checklist of superfi cial behaviours.The sim plicity of the organisation also makes it easy to grasp and recall.

Figure 1 :
Figure 1 : Rating by six assessors of nurses' communication skills before and after training.