Patient Education and Counseling 54 (2004) 197–200

The patients’ written word: a simple communication aid

Thomas Wellsa,∗, Stephen Falka, Paul Dieppeba Bristol Haematology and Oncology Centre, Horfield Road, Bristol BS2 8ED, UK

b Department of Social Medicine, MRC Health Services Research Collaboration, Bristol University, Bristol, UK

Received 28 March 2003; received in revised form 26 June 2003; accepted 6 July 2003

Abstract

Effective doctor–patient communication is an essential part of good medical practice. Question asking is one way for patients to havegreater participation. In this feasibility study, patients were given the opportunity to list questions or discussion topics on a proformabefore seeing the doctor in an oncology outpatient clinic. The items listed were reviewed by the clinic doctor. Eighty-eight of 100 patientsapproached agreed to participate. Biomedical questions (mean 2.0 per patient) predominated over psychosocial ones (mean 0.1 per patient).Possible reasons for this are the study researcher being a doctor, patients attending for test results or that patients consider the format of listingquestions on a proforma inappropriate for psychosocial issues. The listing of questions and discussion topics by patients gave the doctorvaluable information about the patients’ understanding and use of language. While the importance of doctors being adequately trained incommunication skills should not be ignored, this simple intervention has potential for improving the outcome of clinic consultations.© 2003 Published by Elsevier Ireland Ltd.

Keywords: Biomedical; Communication; Listing; Psychosocial; Questions; Topics

1. Introduction

Patients have differing information needs and derive ma-terial from different sources[1–5]. Unvoiced agendas mayexist when patients consult their doctors[6] and doctors mayfail to elicit all of the patients’ reasons for attending[7].Emotional and social agendas are likely to be underrepre-sented[6,8]. Patients have been shown to need to receive in-formation to make treatment choices even if the doctor takesthe lead in making the actual decision, known as psycho-logical autonomy[3]. Since effective doctor–patient com-munication can improve health outcomes such as emotionalhealth, function, symptom resolution and pain control[9], itis important to examine ways in which it might be optimised.

Various communication interventions have been investi-gated. Communication training for health care professionalshas been shown to have a significant effect on their com-munication skills[10,11]. Facilitating patients’ participationin the health professional consultation by coaching them tohave a more active role has also been investigated[12,13],although this did not have a significant effect on consultationoutcome. Another intervention investigated has been ques-tion prompt sheets and problem checklists[12,14–16]. They

∗ Corresponding author. Tel.:+44-117-938-2417;fax: +44-117-928-4409.E-mail address: thomas.wells@ubht.swest.nhs.uk (T. Wells).

have been used in the cancer setting and consist of a list ofsuggested questions for patients to tick and additional freespace in which to add other questions. It was found that thelength of consultations decreased and the number of ques-tions relating to prognosis[14,15], tests and treatment[12]increased, if the question prompt sheets were reviewed bythe doctor. They do, however, have limitations in so muchthat listing questions for patients to tick is patronising andpaternalistic and it might even influence their agendas. Infact, the listing of questions for patients to tick is almost do-ing the opposite of encouraging them to voice their personalagendas.

The intervention used in our study was a proforma onwhich there was the statement “Please list below any ques-tions or items that you would like the doctor to discuss whenyou see him or her in your clinic consultation”, followedby five spaces and additional free space for patients to listquestions and discussion topics. Patients completed the pro-forma before their outpatient clinic consultation at a regionaloncology centre and the questions listed were reviewed bythe doctor in the consultation. The aim of this study wasto investigate the extent to which patients made use of thisfacility to list questions and discussion topics. The num-ber and content of questions listed was evaluated, contentbeing classified as either biomedical or psychosocial. Theimplications of using these listings in routine practice arediscussed.

0738-3991/$ – see front matter © 2003 Published by Elsevier Ireland Ltd.doi:10.1016/S0738-3991(03)00213-1

198 T. Wells et al. / Patient Education and Counseling 54 (2004) 197–200

2. Methods

Four consultant oncologists included their patients inthe study. A letter and patient information sheet was sentto 100 randomly selected patients approximately 2 weeksbefore their clinic appointment. The researcher approachedeach patient on arrival at clinic and asked if they werewilling to participate, even if they did not have any ques-tions or topics to list. He explained the study participationsheet to each participant, showing where on it they shouldlist various details about themselves, such as their age,gender, marital status, occupation, postcode and why theywere attending the clinic. He also asked them to list in thespaces provided any questions or topics that they wouldlike to discuss with the doctor. He mentioned that thedoctor would be given the completed study participationsheet before seeing them in the consultation. The numberof questions listed per patient (biomedical and psychoso-cial) was calculated. In addition, the content of questionswas examined in greater detail to assess specific areascovered.

The number of patients who agreed to participate was88 (21 males, 67 females). Patient age (years) was 60.5(mean), 59 (median) and 27–87 (range). The different can-cer diagnoses were: 56 breast, 9 lung, 4 anus, 4 melanoma,4 thyroid, 3 colon, 3 sarcoma, 2 oesophagus, 1 lip, 1 lym-phoma, 1 mesothelioma. Subgroup analysis was performedfor the following variables to determine whether the number

Table 1Type, examples and number of questions/topics listed

Type of question/topic Examples of questions/topics listed Total number ofquestions/topics

Symptoms Lymphoedema and treatment options. Persistent back pain for 2 weeks. Change ofhormonal treatment and whether it will affect hot flushes? How long can I expect the painto continue and is it normal?

60

Follow up, future testsand management

When does follow up finish? Will further scans, radiotherapy or chemotherapy berequired? How do you know when/if treatment has worked?

45

Prognosis, disease behaviour Chance of reoccurrence? Will be same type of cancer reoccur or a different kind?Doctor’s estimation of my general progress? Can diet help? If the lymph glands showedno evidence of disease, why can the disease flare up after tumour removal?

24

Test results, previoustreatment

Result of recent mammogram? Can the patient see the X-rays and/or scans? Discuss theresults of previous biopsies and whether they are correct.

15

Research treatments,information sources

Sources of useful and relevant reading matter on cancer. Any research into diets toprevent risk of further cancer? Use of shark cartilage.

14

Drugs Tamoxifen side effects. Is it necessary to take tamoxifen for 5 years? Can I have vitaminsfor my veins? Is it advisable to have a flu jab?

12

Emotional Are anxiety/panic attacks a pattern in cancer patients? Worry about cancer returning.Finding the emotional side difficult to cope with. We have both taken a very positiveattitude and always cast any fears and doubts aside.

7

Social Can I drive myself home after treatment? Are long journeys (e.g. to Australia) advisablewhen not feeling 100% well? If I were to go abroad for say 2 years, would I be able tohave notes detailing what has happened to me so that any tests could continue?

3

Family risks Daughter’s risk of cancer, since a few in the family had breast cancer? Does cancer runin families?

2

of questions listed were affected by them:

• gender: male, female;• marital status: married/widowed, single/divorced;• patient age: less than 50, 50–75, over 75 years;• prognosis: good, moderate, poor;• time since first outpatient clinic appointment:≤1, >1 to

≤3, >3 years;• deprivation (Carstairs/Townsend score[17–19]): score 1

or 2, score 3, score 4 or 5.

Non-parametric statistical tests were used for performingthe subgroup analysis. The Mann–Whitney test was used forvariables with two subgroups (gender and marital status).The Kruskal–Wallis test was used for variables with threesubgroups (patient age, prognosis, time since first outpatientclinic appointment and deprivation).

3. Results

The number of questions listed (and the number of pa-tients listing that number of questions) was 0 (23), 1 (13),2 (13), 3 (21), 4 (10), 5 (7) and 7 (1). The average numberof questions listed per patient was 2.1 (mean), 2 (median).Biomedical questions (2.0 mean, 2 median per patient) werelisted more often than psychosocial (0.1 mean, 0 medianper patient). SeeTable 1for examples of different types ofquestions and topics listed.

T. Wells et al. / Patient Education and Counseling 54 (2004) 197–200 199

Table 2Subgroup analysis of number of questions listed per patient for differ-ent variables (gender, patient age, marital status, time since first clinicappointment, prognosis, deprivation)

Number of questions or topicslisted per patient

P

Median Mean 95% CI Range

GenderMale 2.0 1.9 1.1–2.7 0–5Female 2.0 2.2 1.7–2.6 0–7 0.635

Patient age (years)<50 2.0 2.4 1.3–3.5 0–750–75 2.0 2.0 1.5–2.4 0–5>75 3.0 2.2 1.4–2.9 0–5 0.807

Marital statusSingle 2.0 2.6 1.6–3.6 0–7Married 2.0 2.0 1.6–2.4 0–5 0.235

Time since first outpatient clinic appointment (years)≤1 3.0 2.5 1.9–3.2 0–5>1 to ≤3 2.0 1.9 1.1–2.7 0–7>3 1.0 1.5 0.7–2.3 0–5 0.089

PrognosisGood 2.0 2.0 1.5–2.6 0–5Moderate 2.0 2.2 1.1–3.3 0–7Poor 2.0 2.1 1.5–2.7 0–5 0.977

Deprivation (Carstairs/Townsend score)1, 2 2.5 2.3 1.6–3.0 0–53 2.0 2.4 1.7–3.1 0–74, 5 2.0 1.7 1.1–2.4 0–5 0.369

P < 0.05 would indicate a significant difference in the number of questionslisted by the subgroups in the variable analysed.

The number of questions listed was not related to progno-sis, gender, age, marital status or deprivation scores. How-ever, patients tended to put forward less questions as theirdisease history became longer. When grouped accordingto time since first outpatient clinic appointment (0–1, >1and≤3, and >3 years since first appointment), the mediannumber of questions listed per patient were 3, 2 and 1, re-spectively, and this difference approached statistical signif-icance (P = 0.089). SeeTable 2for summary of subgroupanalysis.

4. Discussion and conclusion

The results of this feasibility study indicate that a largeproportion of patients choose to use the facility of list-ing questions and topics on a proforma before consultation.Study recruitment was good, being 88%. Of the 88 partici-pants, 65 listed one or more questions/topics. Nonetheless, itis unknown whether these issues would have been raised inany case. Also, patients might have felt obliged to list itemssimply because the doctor was indirectly asking them to doso, the listing reflecting some social desirability. Althoughit did not reach theP < 0.05 level of significance, morequestions tended to be written by patients in their first year

of attending the outpatient clinic. This is what one mightexpect, as patients are likely to have more questions in theearly stages following their cancer diagnosis. The other vari-ables evaluated in the subgroup analysis did not appear toaffect the number of questions listed. This could be eitherbecause there is indeed no effect of these variables on thenumber of questions listed or that the power of the study istoo small to detect an effect.

4.1. Discussion

Why were the questions listed predominantly biomedical,when previous research suggests that psychosocial agen-das exist and are likely to be underrepresented? It mightreflect the setting, such as the researcher administering thestudy being a doctor or patients attending for test results.The format of listing questions on a proforma could in it-self be considered by patients to be an inappropriate wayto raise psychosocial issues. Also, psychosocial agendasmight be addressed elsewhere and, even though the major-ity of questions were biomedical, psychosocial issues mightunderlie some of the biomedical questions listed. For ex-ample, when patients list the chance of cancer recurrence,it might uncover their underlying anxiety or that familymembers are finding the whole situation rather difficult. Ifpsychosocial needs are actually not being addressed, othermethods to improve communication should be employed.Patients themselves can be taught to participate morein the medical interview[12,13] and, certainly, trainingcourses for cancer doctors have been shown significantlyto improve key communication skills[10,11]. Doctors’communication skills are of paramount importance, what-ever interventions are used to improve clinic consultationdynamics.

In standard practice, most patients do not routinely bringa list of questions to the clinic. In view of the high uptake ofthe facility to list questions in this study, it can be inferredthat giving patients the opportunity to list questions on adedicated proforma is very different from the doctor sim-ply encouraging patients to bring a list of questions when-ever they attend. Unless it is made clear that it is standardpractice and of utmost importance for the consultation to bepatient-centred, patients might be reluctant to list questionsfor fear of delaying the clinic[2]. Letting patients write downquestions and topics on a proforma before the consultationgives them another opportunity to voice their agendas, if itis not considered a substitute for them raising issues ver-bally in the consultation. Indeed, some patients might evenprefer to write down questions for the doctor to raise dur-ing the consultation, especially if they feel awkward aboutthe particular issue listed. Also, the act of writing questionsdown might mean that patients less frequently forget thingsthat they would like to discuss.

As well as identifying the patients’ concerns, the listingsgave insight into the patients’ use of language and knowl-edge base. Occasionally, patients’ uncertainty of prognosis

200 T. Wells et al. / Patient Education and Counseling 54 (2004) 197–200

was made apparent. One lady, who had received second-linechemotherapy for breast cancer with liver and lung secon-daries, wrote, “Is there a risk that cancer will spread to otherparts of the body or is this prevented by current treatment?”Being able to gauge patients’ understanding and expecta-tions in this way can be valuable for the doctor, enablingappropriate handling of the issue. Also, the doctor did notalways understand what was listed and it was apparent thatpatients were uncertain about what had been raised in pre-vious consultations. The listing was a way of signalling thisbreakdown in communication and it highlighted the fact thatdoctors and patients can misunderstand each other withoutacknowledging it. Doctors and patients need to use appro-priate language which both parties understand and to havemethods of acknowledging that both have understood whathas been discussed.

4.2. Practice implications

Our initial impressions are that the proforma used in ourstudy is an acceptable and simple tool with the potential forimproving patient care. Patients need to put their illness intothe context of their life. The listings should enable patientsto discuss issues in the appropriate language for them, getthe desired level of information, and address their concernsand health needs. The study intervention has the scope foruse in many healthcare settings, although it has only beeninvestigated in the oncology outpatient clinic setting. The ef-fects on consultation dynamics and outcome have not beenassessed and further investigation is warranted to assess theimpact of using this tool in routine clinical practice. Valid-ity is an important consideration, and this issue should beaddressed with further development and assessment of thetool in different study populations and contexts, taking intoconsideration the previously mentioned issue of social de-sirability. Assessable outcomes include patient and doctorsatisfaction, consultation time, medication prescription andconsultation content analysis.

Acknowledgements

Consultants who allowed access to their patients; Re-search Development and Support Unit, United BristolHealthcare Trust, for advice on statistical analyses. TheDepartment of Social Medicine of the University of Bristolis the lead centre of the MRC Health Services ResearchCollaboration.

References

[1] Cassileth BR, Zupkis RV, Sutton-Smith K, March V. Informationand participation preferences among cancer patients. Ann Intern Med1980;92:832–6.

[2] Leydon GM, Boulton M, Moynihan C, Jones A, Mossman J,Boudioni M, et al. Cancer patients’ information needs and infor-mation seeking behaviour: in depth interview study. Br Med J2000;320:909–13.

[3] Sutherland HJ, Llewellyn-Thomas HA, Lockwood GA, Tritchler DL,Till JE. Cancer patients: their desire for information and participationin treatment decisions. J R Soc Med 1989;82:260–3.

[4] Thomas R, Thornton H, Mackay J. Patient information materi-als in oncology: are they needed and do they work? Clin Oncol1999;11:225–31.

[5] Wright V, Hopkins R. Patients’ perceptions of staff in a departmentof rheumatology. Br J Rheumatol 1990;29:374–6.

[6] Barry CA, Bradley CP, Britten N, Stevenson PA, Barber N. Patients’unvoiced agendas in general practice consultations: qualitative study.Br Med J 2000;320:1246–50.

[7] Ford S, Fallowfield L, Lewis S. Can oncologists detect distress in theiroutpatients and how satisfied are they with bad news consultations?Br J Cancer 1994;70:767–70.

[8] Fallowfield U, Ratcliffe D, Jenkins V, Saul J. Psychiatric morbidityand its recognition by doctors in patients with cancer. Br J Cancer2001;84:1011–5.

[9] Stewart MA. Effective physician–patient communication and healthoutcomes: a review. Can Med Assoc J 1995;152:1423–33.

[10] Baile WF, Kudelka AP, Beale EA, Glober GA, Myers EG, GreisingerAJ, et al. Communication skills training in oncology. Descriptionand preliminary outcomes of workshops on breaking bad news andmanaging patient reactions to illness. Cancer 1999;86:887–97.

[11] Fallowfield L, Jenkins V, Farewell V, Saul J, Duffy A, Eves R. Effi-cacy of a Cancer Research UK communication skills training modeltor oncologists: a randomised controlled trial. Lancet 2002;359:650–6.

[12] Brown R, Butow PN, Boyer MJ, Tattersall MH. Promoting patientparticipation in the cancer consultation: evaluation of a prompt sheetand coaching in question-asking. Br J Cancer 1999;80:242–8.

[13] Cegala DJ, Post DM, McClure L. The effects of patient commu-nication skills training on the discourse of older patients during aprimary cart interview. J Am Geriatr Soc 2001;49:1505–11.

[14] Brown RF, Butow PN, Dunn SM, Tattersall MHN. Promoting patientparticipation and shortening cancer consultations: a randomised trial.Br J Cancer 2001;85:1273–9.

[15] Butow PN, Dunn SM, Tattersall MHN, Jones QJ. Patient participationin the cancer consultation: evaluation of a question prompt sheet.Ann Oncol 1994;5:199–204.

[16] Cull A, Stewart M, Altman DG. Assessment of and intervention forpsychosocial problems in routine oncology practice. Br J Cancer1995;72:229–35.

[17] Carstairs V, Morris R. Deprivation and health in Scotland. Aberdeen:Aberdeen University Press; 1991.

[18] Majeed FA, Cook DG, Poloniecki J, Martin D. Using data from the1991 census. Br Med J 1995;310:1511–4.

[19] Townsened P, Phillimore P, Beattie A. Health and deprivation: in-equality and the north. London: Routledge; 1988.