08pm246.

qxd 5/7/1999 3:35 pm Page 341

Palliative Medicine 1999; 13: 341348

A longitudinal evaluation of a communication skills

programme
Susie Wilkinson Head of Caring Services Research, Marie Curie Cancer Care, London, Koren Bailey
Research Assistant, Marie Curie Centre, Liverpool, Judith Aldridge Senior Research Fellow, SPARC,
Department of Social Policy and Social Work, University of Manchester and Anita Roberts Lecturer, Marie
Curie Centre, Liverpool
Abstract: Communication is an essential component of palliative care, but patients and
their families are often dissatisfied with their interactions with health professionals.
Communication difficulties are also a recognized stress factor among health
professionals. Education and training, however, are said to improve communication
skills. A communication skills training programme for 110 nurses has demonstrated
statistically significant improvements in the nurses skills, but no long-term effect was
evaluated. This paper presents the results of the long-term follow-up study. Of the 110
nurses contacted, 20% refused to participate, 45% agreed and 35% did not respond;
33 nurses returned usable data.
The mean length of time since completing the original study was 2.5 years. Since
completion of the original course there was no statistically significant deterioration or
improvement in eight of the nine areas of assessment skills evaluated. In the area of
psychological assessment there was a statistically significant improvement, and overall
the nurses maintained their skills and improved in this area. The results suggest that
over time the nurses became more confident in the emotional areas of care as a result
of the training.
The two key elements of the training were audio-tape recordings and feedback, which
raised self-awareness, and experiential workshops covering ways of handling difficult
situations. Whether a similar training effect could be achieved by a short 35-day course
on communication skills needs addressing, because the integrated training skills course
evaluated here is costly. Preliminary evidence suggests that the 35-day course may
not be as effective, so the increased costs associated with the integrated training skills
course may be wisely spent if it improves the quality of nursepatient interaction as
evidence here suggests.
Key words: communication; education, nursing, continuing; neoplasms; nursing
education research; palliative care
Resum: La communication est une composante essentielle des soins palliatifs mais
il arrive souvent que les patients et les familles ne soient pas satisfaits de leurs
interactions avec les professionnels de sant. Les difficults de communication sont
galement reconnues par les professionnels de sant comme un facteur de stress.
Cependant on dit que lenseignement et lentranement amliorent les capacits de
communication. Un programme dentranement la communication auprs de 110

Address for correspondence: Dr Susie Wilkinson, Head of

Caring Services Research/Senior Lecturer in Palliative Care,
Marie Curie Palliative Care Research and Development Unit,
Department of Oncology, Royal Free Campus, Rowland Hill
Street, London NW3 2PF, UK.
Arnold 1999

02676591(99)PM246OA

08pm246.qxd 5/7/1999 3:35 pm Page 342

342

S Wilkinson et al.
infirmires a dmontr des amliorations significatives dans leurs facults de
communication, mais aucun effet long terme navait t valu. Cet article prsente
les rsultats de lvaluation long terme. Sur les 110 infirmires contactes, 20 % ont
refus de rpondre, 45% ont accept et 35% nont pas rpondu; 33 infirmires ont
envoy des rsultats utilisables. La dure moyenne coule depuis ltude initiale tait
de 2.5 ans. Depuis le cours initial, il ny a pas eu dans 8/9 des domaines de comptence
en communication valus de dtrioration ou damlioration statistiquement
significative. Dans le domaine des comptences psychologiques, il y a eu une
amlioration significative maintenue et augmente dans le temps. Les rsultats
suggrent quavec le temps les infirmires sont devenues plus confiantes dans laspect
motionnel des soins, grce lentranement. Les 2 lments cls de lentranement
ont t dune part des enregistrements audio et leur rtro-contrle, augmentant la prise
de conscience de soi et dautre part des ateliers de gestion du situations difficiles. Quun
effet similaire li lentranement soit obtenu par une session courte dentranement
de 3 5 jours doit tre examin car les sessions intgres denseignement sont ici
coteuses. Un rsultat prliminaire suggre quune session de 3 5 jours nest pas aussi
efficace, si bien que le surcot gnr par une formation en communication intgre
est une sage dpense si elle contribue amliorer les interactions entre patient et
infirmire, ce que les rsultats prliminaires suggrent ici.
Mots cls: communication; enseignement; soins infirmiers continus; noplasies;
recherche en enseignement infirmier; soins palliatifs

Introduction
Poor communication in health care causes a
multitude of problems. Frequently, patients are the
victims of health care professionals communication
difficulties at a time when they are disempowered by
their health problems and the associated stresses.
Conversations with patients facing life-threatening
illnesses are not easy, and the demands on health
care professionals are immense as they endeavour
to deliver care that meets the patients differing
physical and psychological needs.
The social and psychological morbidity associated with a diagnosis of cancer often remains unrecognized and unresolved.1 Clearly, this distress will
have an impact on the illness experience and may
be linked to the increasing patient dissatisfaction
with health care that has been associated with communication deficiencies.2,3
Ford et al.4 examined doctorpatient interactions
within an oncology setting, and found that consultations were clinician dominated rather than

patient centred. The report concluded that

although the patients were well informed about
their diagnosis, prognosis and treatment options,
there was a poor level of psychosocial discussion,
with the patients emotional well-being rarely
pursued. Nurses tend to use a variety of blocking
tactics to prevent patients from divulging their
problems,5 thus keeping patient interactions and
assessments superficial with poor coverage of
psychological needs. Nurses also tend to focus on
physical problems.6 This suggests that care is
planned with little information and an assumption
of need.
Communication skills have been linked to the
assessment of individual patients. Holt7 believes
that the uniqueness of each patient can only be recognized and respected by carrying out a comprehensive assessment that recognizes the patients
agenda. Thompson8 observes that some nurses see
it [assessment] merely as a form to fill in and highlights the lack of commitment at senior nurse level
to value the assessment activity, with students often

08pm246.qxd 5/7/1999 3:35 pm Page 343

A longitudinal evaluation of a communication skills programme 343

performing this important aspect of care without
adequate supervision.
Assessment is possibly the most important stage
of the nursing process as it not only acts as a basis
for all the other stages, but also without an accurate
assessment nurses are unable to plan the good quality, individualized care to which all patients have a
right.9 Furthermore, patients interviewed2 felt that
they often had important information that health
professionals should know in order to help them
provide the most effective and appropriate care,
illustrating how communication is a two-way
process. In addition, patients voiced their dislike of
being regarded as a hospital number or a tumour to
be processed, and wished to be treated with respect
for their personal dignity and needs.
Wilkinson5 questioned the quality of communication training as nurses communication skills do
not seem to have improved over the past 20 years.
This uncertainty regarding the effectiveness of
communication training warranted further investigation. It was noted that nurses who had undertaken a communication skills programme integrated
into a specialist nursing course in cancer care were
better communicators than those who had attended a condensed 35-day workshop.5
Heaven and Maguire10 explored the impact of a
35-day communication course on nurses skills
level and their ability to elicit their patients concerns. Alarmingly, they concluded that, despite an
improvement in the nurses skills, there was no
noticeable development in their ability to identify
their patients concerns. They surmised that basic
skills training is not enough to affect nurses ability
to elicit their patients concerns in the workplace,
and suggested that communication skills courses
need to include input on the handling of emotions,
and to challenge the nurses attitudes and beliefs
about their skills and the consequences of their
actions on patients.
Communication training needs to encompass and
develop the skills required for effective dialogue,
including assessment skills, facilitating skills, techniques for handling difficult questions and raising
self-awareness. These skills can be defined as
behavioural criteria, and can be reliably taught and
assessed.11 Teaching by the use of observation, constructive feedback and rehearsal enables the acquisition of interpersonal skills; experience alone may
well only reinforce bad habits.12 A communication

skills programme which addressed skills, knowledge

and attitudes was designed taking into account previous research findings. The programme was integrated into 6-month post-basic courses in cancer
and palliative care.13
The effects of this communication training programme were evaluated. One-hundred-and-ten
nurses who were undertaking specialist courses in
cancer or palliative care were recruited to the study,
and their communication skills were evaluated by
rating three audio-taped recordings of patient
assessments. The rating scale covered nine key
areas, and had been previously tested for reliability and validity. The results demonstrated that the
pre-course coverage of the key assessment areas was
poor, and that the post-course assessment demonstrated a statistically significant improvement in the
breadth and depth of coverage. The training had
most effect on emotionally laden areas, i.e. the
patients awareness of their diagnosis and prognosis and the psychological impact of the disease, and
illustrated an increased confidence and efficacy,
with 90% of the nurses showing quantifiable evidence of improvement.
While these results were encouraging, there was
a need to establish whether the nurses level of competence was maintained over time.
This study was therefore set up to evaluate the
long-term effects of a communication skills programme on nurses communication skills.

Method
Population
Registered nurses who, between 1991 and 1996, had
undertaken any of the following:
Diploma in Cancer Nursing (ENB 237);
Diploma in Palliative Care (ENB 285);
Marie Curie Advanced Award in Palliative Care,
MCAA (within the Diploma/BSc (Hons) Degree
in Health Studies).
Procedure
Each nurse who met the above criteria was sent a
letter asking whether they would be willing to participate in a follow-up study and were asked to
return, in two stamped and addressed envelopes
(SAE), the completed request form either accepting or declining participation in the study.

08pm246.qxd 5/7/1999 3:35 pm Page 344

344

S Wilkinson et al.

On receipt of the completed form, each consenting nurse was sent an audio-tape cassette with a letter asking them to complete an audio-taped
recording of a patient assessment and to return it
in a SAE. Each consenting nurse received written
feedback on their assessment tape following data
analysis.
Rating of audio-tapes
The rating scale used in the original study13 was
used for the follow-up study. The rating scale consists of nine key areas, each area is scored 03
according to the criteria laid down for each of the
nine key areas in the rating manual, with an overall
score of 027.
Analysis of data
Paired t-tests were used to contrast pre-test with
post-test scores, pre-test with follow-up scores and
post-test with follow-up scores, on individual areas
and on overall communications scores. The Bonferroni correction was applied to take account of the
multiple contrasts carried out (this involved multiplying P values by 3).

Results
Sample
Of the 110 nurses who participated in the original
study, 50 (45%) agreed to participate in the followup study, 22 refused (20%) and 38 did not reply
(35%). It had been a considerable time since some
of the nurses had undertaken their post-registration
course and therefore not all the letters sent out were
received, and were returned return to the sender.
Reasons given by the 22 nurses who refused to take
part in the study, included long-term sickness,
maternity leave, other academic commitments and
a change in their employment profile creating difficulty in undertaking the task.
The 50 nurses who agreed to be involved in the
secondary study were found to be representative of
the original sample. A comparison of demographic
variables and assessment scores between those who
agreed to participate in the follow-up study and
those who refused was undertaken. There were no
statistically significant differences between the two
samples.
Of the 50 nurses who agreed to participate in the

follow-up study, 35 returned tapes (a 70%

response rate). Disappointingly, two returned
tapes were of such poor quality they could not be
assessed, thereby reducing the data set to 33.
To ascertain if the 33 nurses were representative
of the original sample, their demographic data
and assessment scores were compared with the 77
nurses who either refused to participate, did not
reply or who agreed then did not submit a tape. The
mean pre-original course score for the follow-up
group was 10.73 compared to 9.86 for those
who refused to undertake the follow-up. The postoriginal course score for the follow-up group
was 16.27 compared to 15.7 for those who refused
to undertake the follow-up. Although the follow-

Table 1

Descriptive statistics of participating sample

Sex
Male
Female
Marital status
Married
Single
Divorced
Widowed
Separated
Age range 2558 years
Religion
Protestant
Catholic
Other
None
Workplace
Hospital
Hospice
Community
Specialist nurse
Official position
Staff nurse
Community sister
Ward sister
Teacher
Specialist nurse
Qualifications
ENB qualifications:
Yes
No
Palliative care qualifications:
Yes
No
Communication qualification:
Yes
No
ENB 237 qualification:
Yes
No

% (33)

2
31

6
94

20
10
1
1
1

61
30
3
3
3

Mean

36.5 years
15
14
3
1

46
42
9
3

10
15
2
6

30
46
6
18

16
1
9
1
6

49
3
27
3
18

24
9

73
27

17
16

51
49

24
9

73
27

1
32

3
97

08pm246.qxd 5/7/1999 3:35 pm Page 345

A longitudinal evaluation of a communication skills programme 345

up group post-original course scores were slightly
better than for those who refused to take part in
the follow-up study, the differences did not reach
the level of statistical significance. There were
no significant differences between the follow-up
group and the refusers on any demographic
variables.
Follow-up sample characteristics
The demographic data, qualifications and employment details for the sample are illustrated in Table
1. Sixteen (49%) of the nurses were staff nurses and
nine (27%) were ward sisters. Of the remainder, six
were specialist nurses (18%) with one community
sister (3%) and one teacher (3%). The greatest proportion of the sample, 15 (46%), were employed

Table 2

within a hospice, followed by 10 (30%) who were

employed within a hospital. Of the remainder, two
were community nurses (6%) and six (18%) were
specialist nurses. The mean length of time that had
elapsed since qualification was 13.5 years (range of
237 years). The mean length of time since completion of the original course (range 15 years) was
2.9 years.
Follow-up nurses coverage scores
Pre-test scores. Prior to undertaking the original
communication skills training programme, the follow-up nurses had been assessed on how well they
covered the nine key areas of the nursing assessment. Table 2 illustrates the coverage of each key
area and the total mean scores.

Frequency distributions and descriptive statistics for assessment areas in pre-test, post-test and follow-up
Coverage (%)
None Poor

Adequate Good Mean

Median

Standard deviation Range

Pre-test (n = 33)
1. Introduction to nursing assessment
2. Patients understanding of admission
3. Patients awareness of diagnosis
4. Patients history of present illness
5. Patients history of previous illness
6. Physical assessment of patient
7. Social assessment of patient
8. Psychological assessment of patient
9. Closure of nursing assessment
Total

46
27
18
12
58
6
3
33
33

6
33
24
27
27
46
46
55
46

46
30
36
55
15
46
46
9
18

3
9
21
6
0
3
6
3
3

1.06
1.21
1.61
1.55
0.58
1.45
1.55
0.82
0.91
10.7

1
1
2
2
0
1
2
1
1
11

1.03
0.96
1.03
0.79
0.75
0.67
0.67
0.73
0.80
3.73

03
03
03
03
02
03
03
03
03
418

Post-test (n = 33)
1. Introduction to nursing assessment
2. Patients understanding of admission
3. Patients awareness of diagnosis
4. Patients history of present illness
5. Patients history of previous illness
6. Physical assessment of patient
7. Social assessment of patient
8. Psychological assessment of patient
9. Closure of nursing assessment
Total

6
27
3
9
39
9
3
24
27

9
3
0
3
33
15
36
21
12

33
58
36
52
18
49
46
24
46

52
12
61
36
9
27
15
30
15

2.30
1.55
2.55
2.15
0.97
1.94
1.73
1.61
1.48
16.3

3
2
3
2
1
2
2
2
2
18

0.88
1.03
0.67
0.87
0.98
0.90
0.76
1.17
1.06
4.00

03
03
03
03
0-3
03
03
03
03
823

Follow-up (n = 33)
1. Introduction to nursing assessment
2. Patients understanding of admission
3. Patients awareness of diagnosis
4. Patients history of present illness
5. Patients history of previous illness
6. Physical assessment of patient
7. Social assessment of patient
8. Psychological assessment of patient
9. Closure of nursing assessment
Total

15
30
3
15
64
6
21
0
18

15
12
3
9
21
49
36
15
39

36
21
24
21
9
33
36
42
36

33
36
70
55
6
12
6
42
6

1.88
1.64
2.61
2.15
0.58
1.52
1.27
2.27
1.30
15.2

2
2
3
3
0
1
1
2
1
15

1.05
1.27
0.70
1.12
0.90
0.80
0.88
0.72
0.85
3.95

03
03
03
03
03
03
03
03
03
624

08pm246.qxd 5/7/1999 3:35 pm Page 346

346

S Wilkinson et al.

Table 3

Pre-test, post-test and follow-up differences in individual areas of assessment

Assessment
Introduction to nursing assessment
Patients understanding of admission
Patients awareness of diagnosis/condition
Patients history of present illness
Patients history of previous illness
Physical assessment of patient
Social assessment of patient
Psychological assessment of patient
Closure of nursing assessment
Total

Pre-test vs post-test
***
ns
***
**
ns
*
ns
**
*
***

Pre-test vs follow-up
**
ns
***
ns
ns
ns
ns
***
ns
***

Post-test vs follow-up
ns
ns
ns
ns
ns
ns
ns
**
ns
ns

*P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant. Bonferroni correction applied.

The coverage scores were low, particularly in

the areas of psychological assessment. Twentynine nurses (88%) either omitted or poorly
addressed the patients feelings, 16 (46%) did not
introduce themselves or state the purpose of the
assessment, 19 (58%) did not cover the patients
history of previous illnesses and nine nurses (27%)
did not ascertain the patients understanding of
their admission.
Coverage of the patients history of present illness
was of an adequate or good level in 20 (61%) of the
cases and the nurses physical patient assessment
was adequate or good in 16 (49%) of cases, indicating the nurses tendency to focus on patients
physical aspects. Out of a possible 27, the total mean
score for the pre-test was 10.7 (range 418),
demonstrating the quantity of important information that remained undiscussed and a very low level
of assessment skills.
Post-test scores. Table 2 shows the post-test
results; the total mean post-training score was 16.3
(range 823). The coverage score for every key area
improved.
Pre- and post-test scores were compared (Table
3). The nurses performance improved in every area
of assessment, the improvement was statistically significant in the following areas: introduction,
patients awareness of diagnosis, history of present
illness, physical assessment, psychological assessment and closure of nursing assessment. The
largest improvement was in the areas of introduction, and awareness of diagnosis and prognosis.
Follow-up scores. Table 2 documents the nurses
follow-up scores. The total mean score dropped to

15.2 from 16.3; in Table 3, the pre-, post-test and follow-up scores are compared.
There was a statistically significant increase in the
total mean score from pre-test (10.7) to post-test
(16.3) (P < 0.001), and from pre-test (10.7) to follow-up (15.2) (P < 0.001). Between the post-test
and the follow-up there was a drop in the score, but
this was not statistically significant. Thus, from the
total assessment scores, the nurses maintained the
improvement in communication skills that they had
achieved post-test.
From the pre-test to the follow-up scores, there
were improvements in all key areas except patients
history of previous illness which remained the same.
The improvements reached levels of statistical significance in areas of introduction, awareness of
diagnosis (P < 0.001) and psychological assessment
(P < 0.001).
There was no statistically significant relationship
between years since completion of the original
course and the total pre-test, post-test or followup scores. There was, however, a significant
correlation between the post-test physical assessment and years since completion (r = 0.459,
P < 0.05), such that those who completed earlier
tended to have higher post-test scores in this area.
There was no statistically significant relationship
between the demographic and employment data
(i.e. place of work and position and the follow-up
scores). The results of the nurses scores for the follow-up study suggest communication assessment
skills can be taught, and levels of competence maintained.

08pm246.qxd 5/7/1999 3:35 pm Page 347

A longitudinal evaluation of a communication skills programme 347

Discussion
Even though some nurses had completed their
training 5 years previously, overall the nurses maintained their level of competence when assessing
patients. There was a significant improvement in
performance from post-test to follow-up in the psychological assessment but not in any of the other
eight areas. The psychological assessment performance is measured in terms of the nurses ability to
pick up on cues and explore in depth how the
patients illness has affected their life and psychological well-being. Prior to training this was an area
that many nurses stated they found difficult, and the
pre-test results support this. Post-test and follow-up,
the scores had improved, demonstrating a gradual
improvement over time.
It would seem that as time passes after the integrated training the nurses do feel more competent
in assessing the psychological aspects of care,
which was the area in which the nurses performance had the greatest improvement. It is possible
that once nurses feel they have the basic skills to
facilitate a deeper level of the nursepatient interaction, they gradually gain confidence in using these
skills in their clinical practice and integrate them
progressively into their patient relationships.
Forty-six per cent of nurses worked in hospices.
Although the environment in which they work has
been shown to influence nurses communication
skills5 there was no difference in scores between the
places of work, illustrating that, contrary to expectations, the nurses in the hospice environment may
not be any different to those working in other environments. What seems to make the difference is the
attitude, management and communication skills of
the managers. To communicate openly nurses need
to feel secure in the knowledge that if they answer
patients openly and honestly they will not get into
trouble with either their manager or the doctors.5
It is essential that communication strategies
between the multiprofessional team are drawn up
to aid this process. This will also enable patients
communication needs to be met.
It should be remembered that the nurses had chosen to undertake a palliative care/oncology qualification and may have undertaken these courses
whilst quite new to the field. The experience that
they have gained since the course may have augmented and consolidated their training. It is diffi-

cult to generalize the findings of this study as the

small sample was self-selected, however, it did not
differ significantly from the original sample. Therefore, the results do indicate that nurses can be
taught communication skills, and their level of skills
can be maintained over time. It would appear that
the programmes unique feature is that it did not
just focus on communication skills, but also knowledge of cancer and palliative care, nurses attitudes
to cancer, death and dying, and raising nurses
awareness of how they communicate.
Although there is some evidence to suggest nurses who complete this integrated approach to
training are better communicators than those
who completed a 35-day course,13 research needs
to address this issue. The integrated approach of
2630 h is very costly in terms of lecturers and
students time. A 3-day workshop of 18 h, if it
achieved the same improvements, would reduce
costs considerably. The different approaches to
teaching communication skills need to be tested,
only by addressing this issue will knowledge on communication skills training be advanced and patients
be helped to receive optimum care. In conclusion,
this study has demonstrated nurses communication
skills can be improved with training. The effect of
training is maintained over time, and nurses
appear to become more confident in handling emotional issues, but the teaching method is costly. The
results of this study do not concur with a previous
study14 which indicated hospice nurses had difficulty eliciting patients psychological concerns. This
training in contrast appears to have enhanced their
skills in the psychological domain. Although it is
recognized that the patients concerns were not
independently assessed, further data analysis15
indicates that 95% of patients that were involved in
this study were satisfied with their interaction. Only
5% of patients felt they would have liked to have
disclosed further concerns.15
Acknowledgements
This research project has been gratefully supported
by the Marie Curie Cancer Care Palliative Care
Research Committee. The author is a member of
the European Communication Biomed II Project,
European Palliative Care: Ethics and Communication.

08pm246.qxd 5/7/1999 3:35 pm Page 348

348

S Wilkinson et al.

References
1 Maguire P. Improving the recognition and treatment
of affective disorders in cancer patients. In:
Grossman Granville K ed. Recent advances in
psychiatry 7. Edinburgh: Churchill Livingstone, 1992.
2 National Cancer Alliance. Patient-centred cancer
services: What patients say. Oxford: National Cancer
Alliance, 1996.
3 Audit Commission. What seems to be the matter:
communication between hospital and patients. Audit
Commission, National Health Service, Report No.
12, 1993.
4 Ford S, Fallowfield L, Lewis S. Doctorpatient
interactions in oncology. Soc Sci Med 1996; 42:
151119.
5 Wilkinson S. Factors which influence how nurses
communicate with cancer patients. J Adv Nurs 1991;
16: 67788.
6 Booth K, Maguire P, Butterworth T, Hillier V.
Perceived professional support and the use of
blocking behaviours by hospice nurses. J Adv Nurs
1996; 24: 52227.

7 Holt P. Role of questioning in patient assessment. Br

J Nurs 1995; 4: 114546.
8 Thompson D. Too busy for assessment? Nurs J 1990;
4: 35.
9 Crossfield T. How to formulate patient assessment.
Nurs Stand 1989; 3: 45.
10 Heaven C, Maguire P. Training hospice nurses to
elicit patient concerns. J Adv Nurs 1996; 23:
28086.
11 Maguire P. Can communication skills be taught?
Br J Hosp Med 43: 21516.
12 Silverman J. Interpersonal and communication skills
can be taught. Br Med J 1995; 310: 527.
13 Wilkinson S, Roberts A, Aldridge J. Nursepatient
communication in palliative care: an evaluation of a
communication skills programme. Palliat Med 1998;
12: 1322.
14 Heaven C, Maguire P. Disclosure of concerns by
hospice patients and their identification by nurses.
Palliat Med 1997; 11: 28390.
15 Wilkinson S, Bailey K. Cancer patients perceptions
of nurses communication skills. Int J Palliat Nurs
1998; 4: 305.