Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Improving the letters we write: an exploration of doctor–doctor communication in cancer care

Improving the letters we write: an exploration of doctor–doctor communication in cancer care British Journal of Cancer (1999) 80(3/4), 427–437 © 1999 Cancer Research Campaign Article no. bjoc.1998.0374 Improving the letters we write: an exploration of doctorÐdoctor communication in cancer care 1 1,2 3 D McConnell , PN Butow and MHN Tattersall 1 2 3 Medical Psychology Unit and Department of Psychological Medicine, University of Sydney, NSW 2006, Australia; Department of Cancer Medicine, University of Sydney, NSW 2006, Australia Summary Referral and reply letters are common means by which doctors exchange information pertinent to patient care. Twenty-eight semi- structured interviews were conducted exploring the views of oncologists, referring surgeons and general practitioners. Twenty-seven categories of information in referral letters and 32 in reply letters after a consultation were defined. The letters to and from six medical oncologists relating to 20 consecutive new patients were copied, and their content analysed. Oncologists, surgeons and general practitioners Australia wide were surveyed using questionnaires developed on data obtained above. Only four of 27 categories of referral information appear regularly (in . 50%) in referral letters. Oncologists want most to receive information regarding the patient’s medical status, the involvement of other doctors, and any special considerations. Referring surgeons and family doctors identified delay in receiving the consultant’s reply letter as of greatest concern, and insufficient detail as relatively common problems. Reply letters include more information regarding patient history/background than the recipients would like. Referring surgeons and family doctors want information regarding the proposed treatment, expected outcomes, and any psychosocial concerns, yet these items are often omitted. Consultants and referring doctors need to review, and modify their letter writing practices. Keywords: letters; referrals; cancer consultation; multidisciplinary care; communication Optimal patient care hinges at least in part on adequate and timely wished to receive information regarding the prognosis and what the exchange of information between treating doctors (Newton et al, patient had been told, yet less than 20% of letters adequately 1992). The referral and reply letters are the most common means by covered these topics. The more recent study, conducted in which doctors exchange information pertinent to patient care Australia, examined 94 reply letters sent by one oncologist (Tattersall et al, 1995). If these letters meet the respective needs of (Tattersall et al, 1995). A questionnaire was sent to 55 GPs and 53 consultants and referring doctors, discontinuity in care, unnecessary referring specialists who had received a letter from the oncologist repetition of diagnostic tests and poor patient outcomes such as asking them to rate each of 14 items as essential, useful, of little anxiety, dissatisfaction and loss of confidence in medical practitioners use, or of no use. may be avoided (Cummins et al, 1980; McPhee et al, 1984; Hull and The majority of respondents (n 5 95) rated the following items Wosterman, 1986; Nutting et al, 1992; Graham, 1994; Epstein, 1995). as essential: diagnosis, clinical findings, test results, further tests, Few studies have investigated the information content of doctors’ treatment options and recommendations, prognosis, and likely letters and/or information preferences of doctor recipients. benefits and side-effects. Less than 50% of doctors regarded Only one study has examined referral letters in the cancer care details of the patients’ presenting history, drug or social history as setting. In this Australian study a limited audit was made of 103 essential. Content analysis of the reply letters found that they consecutive new patients seen by one radiation oncologist (Graham, usually did not specify prognosis, give recommendations of 1994). Of the 80 letters available, 95% reported the diagnosis, but further tests, or specify the likely side-effects of treatment, and only 56% provided a history of the current illness. Less than half the more commonly than referring doctors desired, included details on referrals detailed clinical findings or included information on past presenting history, past medical, drug and social history. The history, social history, medications and allergies. The author extent to which these findings can be generalized, however, is concluded that relevant and important information was not commu- unknown. The letters analysed were from only one oncologist and nicated in referral letters. criteria ‘presumed ideal’ were used for the content analysis, and to Only two studies have specifically investigated the content of identify doctors’ information preferences. letters from oncologists, and the information preferences of the We have conducted a comprehensive audit of referral and reply recipients. Bado and Williams (1984), in their survey of 73 general letters to and from Australian oncologists and explored their infor- practitioners (GPs), reported that technical topics, such as diag- mation preferences and those of referring doctors (surgeons and nosis, findings on investigation and treatment details, were more GPs). Our objectives were as follows: important than social topics. More than 80% of GPs, however, • to determine the purpose/function and preferred content of referral and reply letters as perceived by oncologists and refer- Received 29 June 1998 ring doctors respectively Revised 10 October 1998 • to obtain a representative view of oncologists concerns with Accepted 20 November 1998 referral letters and referring doctors concerns regarding reply Correspondence to: MHN Tattersall letters 427 428 D McConnell et al Table 1 What oncologists want in most/all cases and what they get in referral letters Content items Medical oncologists Radiation oncologists Actual content (n 5 113) (n 5 43) (n 5 89) % Factor 1 – Patient’s wishes/concerns Mean 5 2.58 Mean 5 3.02 % most/all % most/all How the patient is coping 48.2 63.4 6.7 The patient’s information, involvement and treatment preferences 51.8 64.3 18 Impact of the cancer and treatment on the patient’s work, 37.8 56.1 6.7 leisure and self-care activities Factor 2 – Patient’s background Mean 5 2.61 Mean 5 2.82 %% Family history of cancer 39.3 52.4 3.4 Social history – lifestyle e.g. smoking, drinking 43.2 50 10.1 Social history – employment and home situation 44.2 42.9 15.7 Clinical/findings on examination 57.5 88.1 15.7 Factor 3 – Patient’s medical status Mean 5 3.45 Mean 5 3.43 %% Inter-current medical conditions 85 81 22.5 Past medical history 68.2 69 20.2 Current medication 93.8 83.3 21.3 History of presenting problem 80.4 87.5 82 Factor 4 – Involvement of other doctors and their views Mean 5 3.33 Mean 5 3.67 %% Involvement of other doctors in the case 81.3 90.5 23.6 Referring doctor’s view of his/her continuing 69.9 90.5 19.1 involvement in the case What opinions have been expressed by 69.9 85.7 11.2 other doctors about patient management What the patient has been told 80.5 90.5 13.5 The referring doctor’s thoughts on what may 52.3 73.8 32.6 be appropriate management Any factors possibly mitigating against particular 85 100 5.6 treatments or treatment arrangements Tests/findings on investigation 98.2 100 61.8 Factor 5 – Special considerations Mean 5 3.25 Mean 5 3.31 %% Concerns about psychiatric/social problems 75.9 78 3.4 Concerns about patient compliance 68.8 76.2 2.2 Concerns about patient understanding 67.9 73.8 2.2 Wishes/concerns of the patient’s family 58 75.6 1.1 Need for an interpreter 87.4 78.6 1.1 Information regarding any formal clinical trials the 71.4 69 1.1 patient is on or is eligible for Additional items % Mean % Mean Clearly stated reason for referral 98.2 3.94 97.6 3.9 78.7% Provisional diagnosis 88.3 3.51 97.6 3.9 88.8 Copies of test results/reports/films 94.6 3.85 95.2 3.88 N/A Items are listed in order of factor loading. Discrepancies in which . 75% of both medical & radiation oncologists want an item in most/all cases, but , 25% of a b letters actually contain this item are shown in bold. Denotes a significant difference between mean scores at the level of P , 0.05. Denotes a significant difference between mean scores at the level of P , 0.01. • to determine what information is ‘typically’ contained in 28 semi-structured interviews with doctors were conducted referral letters to oncologists, and their reply letters including seven with oncologists from three Sydney hospitals, ten • to prepare a template of referral and reply letters which may with surgeons and 11 with GPs practising in the Sydney enhance communication between referring doctors and oncolo- Metropolitan area. Two interviews were conducted by telephone gists. with GPs in rural areas. All other interviews were conducted in person. The interviews explored doctors’ views on referral communications with a focus on their information needs. All inter- METHOD views were audiotaped and transcribed. The interview data were analysed using the constant-compara- Stage 1 – qualitative phase tive method proposed by Glaser and Strauss (1967). Put simply, In Stage 1, three medical and three radiation oncologists were this involves coding each unit of meaning (i.e. specific response), invited to participate in an interview and to provide contact details and comparing and contrasting these to identify recurring regular- of their last four new patients, their referring doctors and GPs. An ities and discrete categories. This resulted in the development of invitation to participate was then sent to these doctors. A total of an analytic framework of 27 discrete categories of information for British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 429 Table 2 What surgeons and GPs want in most/all cases and what they get in reply letters Content items Surgeons GPs Actual content (n 5 99) % Factor 1 – History/background Mean 5 2.53 Mean 5 3.03 % most/all cases % most/all cases Lifestyle risk factors 35.6 56.1 68.7 Family history of cancer 40.7 70.4 66.7 History of presenting problem 42.6 73.6 97 Past medical history 24.6 37.9 82.8 Social history 24.1 50.4 69.7 Current medication 55.6 89.8 73.7 Inter-current medical conditions 59.2 69.4 64.6 Restatement of reason for referral 31.5 75.7 6.1 Factor 2 – Psycho-social concerns Mean 5 3.13 Mean 5 3.61 %% Concerns about patient understanding 64.9 84.3 6.1 Concerns about psychiatric/social problems 59.4 83.1 1 Concerns about patient compliance 68 88.8 2 Patient’s wishes/expectations regarding information disclosure, 66.7 86.1 26.3 decision making/treatment (3) Impact of cancer and/or treatment on patient’s work, 53.7 79.6 6.1 leisure and self-care activities Likely prognosis (5) 81.5 95.4 31.3 How patient is coping/feeling about diagnosis/prognosis/treatment 68.5 87.9 16.2 Factor 3 – Examination and investigation findings Mean 5 3.66 Mean 5 3.92 %% Tests/findings on investigation 92.6 98.1 41.4 Clinical/findings on examination 74.1 95.3 89.9 Treatment recommendation 94.4 100 85.9 Diagnosis/provisional diagnosis 86.8 100 96 Factor 4 – Future management/expectations Mean 5 2.90 Mean 5 3.71 %% Likely short- and long-term side-effects 58.4 93.4 16.2 Suggestions for management of side-effects 43.6 91.5 5.1 Indicators for unscheduled review by the oncologist 52.8 85.8 8.1 Aim of treatment e.g. curative or palliative (5) 81.1 97.2 40.4 Intention of the oncologist to contact the referring Dr/GP in 54.7 87.8 51.5 the future (4) Factor 5 – Treatment/management plan Mean 5 3.57 Mean 5 3.78 %% The oncologist’s follow-up plan 90.5 98.1 67.7 Involvement of other doctors in the case 80.8 89.7 32.3 Rationale for recommended treatment (3) 79.2 91.6 66.7 Arrangements made for treatment, i.e. where and when 77.4 85.9 33.3 What the patient has been told 88.4 92.4 49.5 Anything specific the oncologist would like the referring Dr/GP 92.4 99.1 14.1 to do. Treatment options 84.9 94.4 31.3 Additional item % Mean % Mean Information regarding any formal clinical trial discussed with 75.5 3.30 85 3.55 10.1 the patient ( ) Indicates that the item also loaded on the factor shown in brackets. Items are listed in order of factor loading. Percentage figures shown in bold highlight a b discrepancies between actual content and preferences of . 50%. 5 P , 0.01; 5 P , 0.001. referral letters and 32 for consultation reply letters (Tables 1 and patients’ medical files were then traced, and referral and reply 2). Common problems encountered in communication between letters photocopied. During data collection, 21 files were not avail- doctors were identified. This analytic framework was used in able and an additional ten referral letters were absent from files. A Stage 2 to analyse the content of referral and reply letters and sample of 89 referral letters and 99 consultation reply letters was provided the basis for the development of questionnaires used in therefore obtained. Stage 3 to survey each group of doctors. Most of the referral letters (77%) were from surgeons or other medical specialists, and 93% were outpatient referrals. The content of each letter using the analytic framework developed in Stage 1 Stage 2 – Content analysis of referral and reply letters was determined by simply noting whether each item of informa- Six medical oncologists from two Sydney hospitals were asked tion was present. The first and third author each analysed a random to provide a list of their last 20 consecutive new patients. The selection of ten letters. Agreement between raters was moderately © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 430 D McConnell et al Table 3 Sample characteristics: Stage 3 to the patient’s situation, and the reason for referral, (b) contributes to assessment by reducing the likelihood of relevant information Characteristics Surgeons GPs Oncologists being overlooked, (c) improves efficiency and quality of care by reducing unnecessary duplication of tests, and providing a focus Sample size n 5 55 n 5 108 n 5 156 Gender for history taking, and (d) provides the groundwork for ongoing Male 54 (98%) 65 (60%) 133 (85%) care and communication. Oncologists reported that missing Female 1 (2%) 43 (40%) 23 (15%) reports or tests results and insufficient detail in the referral letter Years of experience were the most frequent concerns and these were more problematic Mean 19.74 16.59 12.56 than any other (P 5, 0.05). Range 4–40 2–50 0–39 Speciality N/A N/A General surgeon 32 (57%) Actual vs preferred content of referral letters Other surgeon 23 (43%) Twenty-seven categories of information sought in referral letters Average number of cancer Data not Data not were identified in Stage 1. The questionnaire explored oncolo- patients per year collected 2 (2%) collected , 1 33 (31%) gists’ preferences for information in new patients referral letters, 1–5 27 (25%) and respondents indicated on a four-point scale the proportion of 6–10 45 (42%) cases (none, some, most, or all) in which they would like to . 10 1 receive each of the 27 items of information. The aim was to iden- tify ‘in-general’ preferences and priorities for information and to examine current practice in light of these. high at 86%, supporting the reliability and utility of the informa- To identify groups/clusters of items, a factor analysis was tion categories. Upon completion of the coding, nine randomly undertaken. With oblique rotation, a five-factor resolution selected referral and reply letters were recoded to examine intra- emerged, accounting for 51.7% of the variance. Two items, rater reliability. A high level of intra-rater agreement was obtained ‘reason for referral’ and ‘provisional diagnosis’ did not load on at 98%. any of the factors above 0.325 and were therefore considered sepa- rately in subsequent analyses. Table 1 shows the distribution of the 25 items composing the five factors, the percentage of medical and Stage 3 – survey radiation oncologists wanting each item in most/all cases, and the In Stage 3, questionnaires for oncologists, surgeons and GPs were proportion of letters analysed in Stage 2 in which each item was developed based on data obtained in Stage 1 (Appendix 1). present. Oncologists were asked to indicate (a) their preferences for 27 It is evident that a discrepancy exists between information items of information in a referral letter, (b) the frequency with contained and information desired in referral letters. Only four out which they encountered seven common difficulties in referral of 27 items appear regularly (i.e. in more than 50%) of referral communications and (c) if and when a telephone call was preferred letters, namely, the provisional diagnosis, history of the presenting to a letter. Mirroring this, surgeons and GPs were asked (a) their problem, clearly stated reason for referral and findings on investi- preferences for 32 items of information in letters of reply, (b) the gation. On these four items only, referral letters appear to meet frequency of five common problems in reply letters, and (c) when a oncologists’ information needs/preferences. telephone call is preferred to a letter. The questionnaires were Seven items of information wanted by more than 75% of piloted with three oncologists, surgeons and GPs to ensure clarity medical and radiation oncologists in most or all cases were docu- in wording and format. The resulting questionnaire was sent to all mented in less than 25% of letters. Specifically these items are: members of the Medical Oncology Group of the Royal (1) inter-current medical conditions, (2) current medication, (3) Australasian College of Physicians (n 5 148), and all surgeons (n involvement of other doctors in the patient’s care, (4) what the 5 84) and radiation oncologists (n 5 56) who are members of the patient has been told, (5) any factors possibly mitigating against Clinical Oncological Society of Australia (COSA). The sample of particular treatments, (6) concerns about psychiatric/social prob- 200 GPs was drawn from the Directory of Members of the Royal lems, and (7) need for an interpreter. Australian College of General Practitioners which lists almost 10 In interviews and surveys, oncologists identified circumstances 000 members. The sample of GPs was selected using a randomized in which their information needs/preferences may very. Several block design to ensure a representative proportion from each State variables relating to individual patient characteristics and the and Territory. In total, 113 medical oncologists, 43 radiation oncol- nature of the referral were identified. These variables include: (1) ogists, 55 surgeons and 108 GPs returned completed questionnaires whether the patient is an in-patient or out-patient, (2) whether the representing a 76%, 77%, 65% and 54% response rate respectively. doctors interact in a multi-disciplinary clinic, (3) whether the It was not possible to establish the existence of bias introduced by patient is referred preoperatively or post-operatively, (4) whether these response rates which were rather low in the latter two groups. the cancer problem is simple or complex, (5) how well the refer- Some demographic details are presented in Table 3. ring doctor and oncologist know each other, and (6) whether there are significant psycho-social concerns about the patient. Examining how these variables may affect information needs was RESULTS beyond the scope of this study, and they are not allowed for in the presentation of preferences which follows. The referral letter – views of oncologists Analysis of interview data and responses to the survey question Perceived problems with referral letters concerning the function of the referral letter identified four Oncologists interviewed in Stage 1 identified seven concerns common themes. The letter (a) provides background information with referral letters (Table 4). In the questionnaire, we asked British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 431 Table 4 Perceived frequency in which each problem with referral letters occurs (n 5 156) Problems – in descending order from most to Mean [95% CI] [1 5 always, 7 5 never] least frequent Missing reports/test results – i.e. pathology, 3.13 (2.94–3.32) X-ray films, operation report Insufficient information and detail in the 3.46 (3.26–3.66) referral letter No referral letter received prior to or at the time 4.05 (3.82–4.27) of the consultation Hand-written referral letters which are difficult 4.19 (3.98–4.41) or impossible to read Unclearly specified reason for referral 4.89 (4.64–5.14) No referral letter received at all 5.02 (4.78–5.26) Unnecessary information in the referral letter 5.70 (5.50–5.89) Table 5 Perceived frequency of problems with reply letters Mean (95% CI) [1 5 Always, 7 5 Never] Perceived problems Surgeons GPs 1. Reply letters arriving late – not promptly 4.3774 (3.934–4.821) 3.6944 (3.412–3.977) 2. Unnecessary information in the reply letter 4.7170 (4.293–5.140) 5.9907 (5.729–6.252) 3. Insufficient information in the reply letter 5.3396 (4.985–5.694) 4.5648 (3.460–5.669) 4. No reply letter received at all 5.6226 (5.210–6.035) 4.6262 (4.328–4.925) 5. Letters that are too technical and 6.1887 (5.887–6.490) 5.8333 (5.577–6.090) consequently difficult to comprehend oncologists to indicate on a seven-point scale the frequency with consultation, (2) when there is sensitive information to convey, which each of these seven problems occur (from always to never), e.g. if the patient is dissatisfied with other doctors or their manage- and then to identify and rank the three that are most problematic. ment to date, (3) if there are personality or psychological issues Mean scores with 95% confidence intervals were computed. that may affect compliance with treatment recommendations, and Oncologists perceive that missing reports or test results and insuf- (4) if the problem is complex and difficult to relate in a letter and ficient detail in the referral letter occur significantly more often multiple opinions have been sought. than any other problem. These concerns were perceived to be significantly more problematic than any other (P , 0.05). The reply letter – views of referring surgeons and GPs Comparison of medical and radiation oncologists Actual vs preferred content of post-consultation reply letters Figures in Table 1 suggest that radiation oncologists want more Thirty-two categories of information were identified in Stage 1 as information than medical oncologists do in most categories. To components of post-consultation reply letters from oncologists. In statistically explore this finding, the mean score of items in each Stage 2, the actual content of the sample of post-consultation reply factor (where 1 5 in no cases and 4 5 in all cases) were computed letters from radiation and medical oncologists were analysed, and separately for each specialty group and compared using t-tests for in Stage 3, preferences of surgeons and GPs for these items of independent samples. Radiation oncologists on average want more information were sought. Surgeons and GPs indicated on a four- information than medical oncologists concerning patients’ point scale the proportion of cases (none, some, most, or all) in wishes/concerns (P , 0.05) and the involvement of other doctors which they liked to receive letters covering each of the 32 items of in the case (P , 0.01). information identified in Stage 1. Our aim was to identify ‘in- Both medical and radiation oncologists primarily want informa- general’ preferences and priorities for letter content, and then to tion regarding the patient’s medical status, the involvement of evaluate a sample of reply letters with reference to these. other doctors and special considerations. Information concerning To identify groups of related items, a factor analysis was the patient’s wishes/concerns and the patient’s history/background conducted using the data from the survey of referring surgeons and appear to be of secondary importance. GPs. With varimax rotation, a five-factor resolution was obtained accounting for 48.4% of the variance. One item failed to load on When would oncologists like the referring doctor to phone any factor above 0.325 and was therefore analysed separately, them? namely, ‘information regarding any formal clinical trial discussed Most oncologists (73%) indicated that they would like the refer- with the patient’. The five groups, and items loading are shown in ring doctor to phone them (1) when the patient needs an urgent Table 2. Also shown is the percentage of surgeons and GPs © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 432 D McConnell et al Table 6 Information prompt sheets Referral letters ® © Reason for referral ® © Provisional diagnosis ® © Succinct history of the problem ® © Relevant information on patient’s medical status – current medications, inter-current medical conditions and relevant past medical history ® Clinical/findings on examination ® © Information on tests performed and results ® Patient’s wishes and concerns, e.g. how the patient is coping, and their information, involvement and treatment preferences ® © What the patient has been told ® © Involvement of other doctors; what role the referring doctor expects to play; other opinions on management ® © Any factors possibly mitigating against particular treatments or treatment arrangements ® © Special considerations, e.g. psychiatric/social problems, concerns regarding compliance or patient understanding, need for an interpreter, and any concerns/wishes of patient’s family ® © Copies of relevant test results/reports Reply Letters . Restatement of reason for referral . History of presenting problem, family history of cancer, current medication, intercurrent medical conditions ! . Clinical findings on examination; tests/findings on investigation ! . Diagnosis and likely prognosis ! . Treatment options, treatment recommendation with rationale, treatment aim ! . Patient’s wishes and expectations, and how he/she is coping ! . Psycho-social concerns, e.g. patient understanding, psychiatric/social problems ! . Management plan – arrangements, follow-up, and involvement of other doctors . Likely short- and long-term side-effects, and suggestions for the management of these ! . What the patient has been told ! . How and when to contact the oncologist/consultant ® 5 Radiation oncologist, © 5 Medical oncologist, ! 5 Surgeon, . 5 GP. wanting each items in most or all cases, and the percentage of Perceived problems with reply letters reply letters including each item. Five potential problems with reply letters were identified in Stage These data suggest that oncologists’ letters do not provide all 1 interviews (see Table 4). The surgeons and GPs surveyed indi- the information surgeons and GPs want. Oncologists’ letters cated on a seven-point scale how often they perceive that each commonly provide details on examination and investigation find- problem occurs, and identified and ranked the three that are most ings (factor 3), and these items are those most often desired by problematic. Mean scores and 95% confidence intervals were surgeons and GPs. However, the majority of surgeons and GPs computed for each identified problem. Both surgeons and GPs want details of the treatment/management plan (factor 5), future perceive that delay in receiving the reply letter is the most management/expectations (factor 4), and psycho-social concerns frequently occurring problem, and the problem which is of most (factor 2), yet these items are rarely mentioned in letters. concern to them. Superfluous information in the reply letter is Oncologists’ letters also frequently detail the patient’s back- perceived by surgeons to be the next most common problem. GPs’ ground/history (factor 1), which make up six of the ten most however, perceive this to be the least common problem. common items in reply letters. These items, however, are those The preferences of surgeons and GPs for information least often desired by referring surgeons and GPs. The data in Table 2 suggest that the information needs/preferences Several circumstances influencing referring doctor’s informa- of referring surgeons and GPs differ. To test this observation, the tion preferences were identified. These include: (1) how well the mean score of items in each factor (where 1 5 in no cases, and 4 5 referring doctor knows the oncologist, (2) whether there are in all cases) for surgeons and GPs were computed and compared routine clinical meetings between the referring doctor and oncolo- using t-tests for independent samples. The results indicate that GPs gist, (3) the reason for referral – e.g. for second opinion or to take on average want more information than surgeons in every cate- over patient management, (4) whether the patient consultation is gory. Both surgeons and GPs place highest priority on receiving pre- or post-surgery, (5) whether the patient is an in-patient or out- details of the examination and investigation findings, and the patient, (6) whether the cancer is rare or common, and (7) whether proposed treatment/management plan. the treatment recommended is standard or not. However, exam- ining how these variables may affect information needs/prefer- When would surgeons and GPs like the oncologist to phone ences of referring specialists and GPs was beyond the scope of this them? study and they are not allowed for in the presentation of prefer- Sixty per cent of surgeons and 78% of GPs identified circum- ences which follows. stances in which a reply letter is insufficient and a phone call British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 433 desirable. Specific circumstances in which referring doctors would has been told, the treatment options, aim of treatment and likely like the oncologist to phone them are (1) when urgent issues arise, prognosis, the involvement of other doctors in the case, and (2) when the treatment proposed is unconventional, and (3) when anything specific the oncologist would like the referring doctor/GP the oncologist is uncertain about the preferred management. A to do. Previous studies have also identified the absence of informa- telephone call is also favoured when (4) divergent views exist on tion on prognosis and what the patient has been told as significant treatment approach and (5) if the treatment recommendation is gaps in the information content of ‘typical’ reply letters. different to that which the referring doctor thought appropriate at It is common practice for oncologists to send GPs a copy of the the time of referral. reply letter to the referring surgeon without alteration. Previous studies have either looked at the information needs of GPs alone, or grouped them together with referring specialists. This study DISCUSSION compared the information preferences of surgeons and GPs, and Doctors write many referral letters either to clinical colleagues or our findings suggest that one reply letter may not adequately meet to diagnostic service providers. Specialist physicians write letters the needs of both. Information preferences appear to be the same, in reply to referring doctors after new patient consultations or with both surgeons and GPs wanting information concerning follow-up visits, and to clinicians caring for patients at home examination and investigation findings most, and information following discharge from hospital. Previous studies suggest the regarding patient history/background least. However, the results of content, legibility, speed of receipt and relevance of doctors’ this study indicate that GPs want significantly more information letters are often deficient and/or do not meet expectations. We than surgeons in every category. These results may explain the have conducted an information audit of referral and reply letters, differences between surgeons and GPs in their perceptions of prob- interviewed and surveyed a sample of referring doctors and oncol- lems with reply letters. Superfluous information is perceived by ogists concerning their preferences and experience with doctors’ surgeons to be the second most common problem with reply letters. The results of this study suggest the need for doctors to letters. GPs, however, perceive this to be the least frequently review, and modify their letter writing practices. occurring problem, if in fact a problem at all. We found that referral letters typically include a statement of the reason for referral, some history of the problem, a provisional Implications for practice diagnosis and description of the findings on investigation. Whilst these items are among the ‘most wanted’, oncologists in this study The findings of this study raise doubts as to whether referral and have clearly articulated a ‘wish’ for a range of additional items of reply letters fulfill their perceived functions. Modifying letter information. At the top of oncologists’ ‘wish list’ is information writing practices may be a relatively simple and effective means concerning the patient’s medical status, the involvement of other of improving doctor–doctor communication and hence, patient doctors and any special considerations. Many oncologists also understanding and outcomes. Referring doctors could improve prefer letters that outline the patient’s history and their wishes and communication between themselves and medical and radiation concerns, but this information appears to be of secondary impor- oncologists by ensuring that available test results/reports accom- tance presumably because these items would be sought during pany the referral letter, by mailing the referral letter to the oncolo- history taking. Radiation oncologists appear to want more infor- gist prior to the consultation and giving a copy to the patient. An mation in referral letters than medical oncologists, particularly in information prompt sheet for referral letters and letters of reply is the areas of patients’ wishes/concerns and the involvement of provided in Table 6. other doctors. Given the significant discrepancy between informa- Medical and radiation oncologists could take several steps to tion desired and information contained, it is not surprising that improve communication with referring surgeons and GPs. Letters oncologists perceive that insufficient information and detail is one should be sent soon after the consultation, since delay in receiving of the two most frequently occurring problems with referral letters. the reply letter is a major concern of both surgeons and GPs. Post-consultation reply letters from oncologists are not meeting Oncologists’ letters should not recount all aspects of the patient the information preferences of referring surgeons and GPs. From history. However, these letters should document the results of the letter writer’s perspective, the reply letter also functions as examination and investigations, the treatment options and a consultation record. Kamien (1995) has highlighted this proposed management plan, state the prognosis and what the dichotomy of purpose, and argued that it must be resolved in the patient has been told, and outline any psycho-social concerns. interests of good communication. Should we write two letters, one Although a case can be made for writing two letters, one for a that is filed in the notes as a record of the consultation, and the referring surgeon (if relevant) that is short and succinct, and one second that is prepared specifically to inform the referring doctor for GPs that is more comprehensive, this is clearly not practical. and meet their information needs? For GPs’ standard information sheets may be included with the Our results confirm previous findings. Tattersall et al (1995) reply letter concerning the cancer type, potential side-effects of the concluded that reply letters, more often than is desired, contain treatment proposed and recommendations for their management. information concerning patient history. This study confirmed that More than 90% of GPs want this information and less than 20% of items of information concerning patient history/background are oncologist reply letters currently provide any of these details. among the most common items in reply letters, but are the least desired. Surgeons and GPs prefer details concerning the treat- Future research ment/management plan, future management/expectations and psycho-social concerns, yet these are rarely provided in reply Future research should examine how the information needs/prefer- letters. There were several items desired by surgeons and GPs in ences of oncologists and referring doctors may vary with the more than 80% of cases, but included in less than 50% of letters. circumstances identified in this study. Such research will permit These were findings on examination, details of what the patient doctors to better predict and tailor their letters to referring and © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 434 D McConnell et al Epstein RM (1995) Communication between primary care physicians and other doctors. In addition, referral and reply letters, which incorpo- consultants. Arch Family Med 4: 403–409 rate the recommendations of this study, should be evaluated to Glaser BG and Strauss AL (1967) The Discovery of Grounded Theory. Strategies for determine whether they result in increased satisfaction on the part Qualitative Research. Aldine: New York of recipients, whether they fulfill their perceived functions as iden- Graham PH (1994) Improving communication with specialists. The case of an tified in this study and whether they result in better patient oncology clinic. Med J Aust 160: 625–627 Hull FM and Westerman RF (1986) Referral to medical outpatients department at outcomes. teaching hospitals in Birmingham and Amsterdam. Br Med J 293: 311–314 Kamien M (1995) Writing to referring doctors. Aust NZ J Med 25: 463–464 ACKNOWLEDGEMENTS McPhee SJ, Lo B, Saika GY and Meltzer R (1984) How good is communication between primary care physicians and subspecialty consultants? Arch Int Med This study is supported by a NSW Cancer Council Patient Care 144: 1265–1268 Award. Newton J, Eccles M and Hutchinson A (1992) Communication between general practitioners and consultants: what should their letters contain? Br Med J 304: REFERENCES 821–824 Nutting PA, Franks P and Clancy CM (1992) Referral and consultation in primary Bado W and Williams CJ (1984) Usefulness of letters from hospitals to general care: do we understand what we are doing? J Family Pract 35: 21–23 practitioners. Br Med J 288: 1813–1814 Tattersall MHN, Griffin A, Dunn SM, Monaghan H, Scatchard K and Butow PN Cummins RO, Smith RW and Inui TS (1980) Communication failure in primary (1995) Writing to referring doctors after a new patient consultation. What is care. Failure of consultants to provide follow-up information. JAMA 243: wanted and what was contained in letters from one medical oncologist? Aust 1650–1652 NZ J Med 25: 479–482 APPENDIX 1 The initial patient consultation with a medical or radiation oncologist: doctor and patient information preferences QUESTIONNAIRE FOR ONCOLOGISTS What information do onocologists want to receive with a new patient referral? What concerns to oncologists have about referral letters? What are the views of oncologists about providing patients with a post-consultation letter? This questionnaire is primarily concerned with these three questions. Please take the time (approximately 15 minutes) to fill it in. Your views are important in order to obtain a representative view of oncologists. If you have any questions about this project, please contact Mr David McConnell on (02) 9515 8160. Your answers will remain strictly confidential. Thank you in advance for your participation. Part A – Treatment decision making and working with other doctors 1) In your opinion, how should treatment decisions be made? Please tick the statement which best describes your opinion (please tick one box only) n n The doctor should make the decisions based on what he/she determines to be the best treatment for the cancer n n The doctor should make the decisions but consider the patient’s priorities and quality of life n n The patient and the doctor should make the decisions together n n The patient should make the decisions, but consider the doctor’s opinion n n The patient should make the decisions using all they know or learn about their treatment options 2) Please indicate the extent to which you agree/disagree with each of the following statements by circling the number which best repre- sents your view (1 5 strongly agree, 7 5 strongly disagree) Generally speaking, for patients who may see other doctors… Strongly Strongly Agree Disagree A. Oncologists should try to ensure the information 1 2 3 4 5 6 7 they give to patients is compatible with that likely to be given by other doctors B. Oncologists should consider the views of the 1 2 3 4 5 6 7 patient’s GP in determining the treatment plan C. Oncologists should consider the views of doctors 1 2 3 4 5 6 7 from other specialities in determining the treatment plan D. Oncologists should share follow-up with doctors 1 2 3 4 5 6 7 from other specialities E. Oncologists should share follow-up with the patient’s GP 1 2 3 4 5 6 7 F. A patient’s cancer care should be jointly managed 1 2 3 4 5 6 7 by the oncologist and the GP G. A patient’s cancer care should be jointly 1 2 3 4 5 6 7 managed by the oncologist and doctors from other specialities H. A patient should be referred to an oncologist prior to surgery 1 2 3 4 5 6 7 British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 435 Part B – Information accompanying referrals 3) In what proportion of cases would you like to receive each item of information listed below in a referral letter? If you tick most or some for any item of information, please specify the circumstances in which you want that information from the referring doctor. Items of information With all With With With no Please use this space to specify referrals most some referrals referrals referrals (specify) (specify) 1. Patient’s social history, n n n n n n n n e.g. employment, home situation 2. Reason for referral n n n n n n n n 3. History of presenting problem n n n n n n n n 4. Family history of cancer n n n n n n n n 5. Social history – lifestyle, n n n n n n n n e.g. smoking, drinking 6. Past medical history – unrelated to n n n n n n n n the presenting problem 7. Inter current medical conditions – physical n n n n n n n n & psychiatric 8. Current medication n n n n n n n n 9. Clinical findings: results of physical n n n n n n n n examination 10. What tests have been done or n n n n n n n n arranged by the referring doctor & a summary of the main findings 11. Diagnosis/provisional diagnosis n n n n n n n n 12. Referring doctor’s thoughts on what n n n n n n n n may be appropriate management 13. What other opinions have been n n n n n n n n expressed by other doctors about patient management 14. Any factors possibly mitigating n n n n n n n n against certain treatments or treatment arrangements – medical, psycho-social, or demographic 15. Referring doctor’s view of his/her n n n n n n n n continuing involvement in the case 16. Involvement of other doctors in the case n n n n n n n n 17. What the patient has been told regarding n n n n n n n n diagnosis, prognosis, treatment options 18. The patient’s wishes, expectations or n n n n n n n n concerns regarding information disclosure, decision making, treatment 19. How the patient is coping and/or feeling n n n n n n n n about their diagnosis, prognosis or treatment 20. Impact of the cancer & its treatment on n n n n n n n n the patient’s work, leisure and self care activities 21. Any concerns about how much the n n n n n n n n patient understands 22. Any concerns about psychiatric and/or n n n n n n n n social problems 23. Any concerns about patient compliance n n n n n n n n willingness to accept advice © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 436 D McConnell et al Items of information With all With With With no Please use this space to specify referrals most some referrals referrals referrals (specify) (specify) 24. Whether an interpreter is required for the n n n n n n n n consultation [if the patient has difficulty speaking English] 25. Information regarding any formal n n n n n n n n clinical trials the patient is on, has been offered, or is eligible for 26. Any wishes/concerns of the patient’s n n n n n n n n family, e.g. about the disclosure of information to the patient 27. Copies of test results, n n n n n n n n e.g. pathology report, X-ray films Would you like to receive any other information from the referring doctor? If so, please specify on the back of this page. 4) Is there any information you would prefer to receive over the phone, or circumstances in which you would like the referring doctor to phone you? n n Yes n n No If yes, please specify. 5) How is the information you receive from the referring doctor helpful? What purpose does it serve? 6) i. You may have experienced the following problems with referral letters. Please circle the number which best represents how often each occurs. (1 5 always, 7 5 never). Always Never A. Missing reports/test results – i.e. pathology, 1 2 3 4 5 6 7 X-ray films, operation report etc. B. Hand-written referral letters which are difficult 1 2 3 4 5 6 7 or impossible to read C. Unclearly specified reason for referral 1 2 3 4 5 6 7 D. Insufficient information and detail in the referral letter 1 2 3 4 5 6 7 E. Unnecessary information in the referral letter 1 2 3 4 5 6 7 F. No referral letter received prior to or at the time of consultation 1 2 3 4 5 6 7 G. No referral letter received at all 1 2 3 4 5 6 7 Please list any other concerns you may have and indicate how often each occurs 12 3 4 5 6 7 12 3 4 5 6 7 7) ii. From the list above, which 3 concerns about referral letters are most problematic? Please list and rank these with 1 being the most problematic. 8) When would you ideally like to receive the referral letter? n n Prior to the patient consultation n n At the time of the patient consultation n n It doesn’t matter 9) In what format would you prefer the referral letter to be written? n n In narrative format n n In point form n n It doesn’t matter British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 437 10) When a patient is not referred by their GP, how often do you practice each of the following activities? (1 5 always, 7 5 never) Always Never i. Send the GP a copy of the letter written to 1 2 3 4 5 6 7 the referring doctor ii. Write an additional letter to the GP 1 2 3 4 5 6 7 iii. Send the GP a copy of the letter written to 1 2 3 4 5 6 7 the referring doctor – with an additional post-script iv. Send the GP a copy of the letter addressed 1 2 3 4 5 6 7 to the referring doctor, but written with the GP in mind Part C – Patient information Several studies suggest that patients have difficulty remembering information conveyed in their initial consultation. We would like to obtain your views on 3 strategies which may address this problem. 11) A. Do you think patients should be offered a copy of the letter written to the referring doctor? n n Yes n n No n n It depends Please explain: B. Do you think patients should be offered an individualized/personal letter as a follow-up to their consultation with you? n n Yes n n No n n It depends Please explain: C. Do you think patients should be offered an audiotaped recording of their consultation with you? n n Yes n n No n n It depends Please explain: 12) Which of the above strategies for providing information do you most prefer? n n a copy of the letter written to the referring doctor n n an individualized/personal letter following their consultation n n an audiotaped recording of their consultation n n None of the above 13) In your opinion, are there any ‘better’ strategies (better than those listed above) to ensure that patients are adequately informed? n n Yes n n No If yes, please specify: 14) In what proportion of cases do you practice each of the following activities? Please tick. In all In most In some In no cases cases cases cases i. Dictate your letter to the referring doctor in front of the patient n n n n n n n n ii. Offer patients a copy of the letter written to the referring doctor n n n n n n n n iii. Offer patients an individualized/personal n n n n n n n n letter after the consultation iv. Offer patients an audiotaped recording of the consultation n n n n n n n n v. Offer patients general information booklets n n n n n n n n 15) Would you like to make any further comments about any of the issues raised in this questionnaire? Personal Details: 16) Your sex n n Male n n Female 17) Your speciality n n Medical Oncology n n Radiation Oncology 18) How would you best describe your current position? n n University appointment n n Visiting Medical Officer n n Staff specialist n n Private practitioner n n Other 19) In what institution is your main practice? n n Private hospital n n Teaching hospital n n District hospital n n Other 20) For how many years have you been a practising oncologist? years. © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png British Journal of Cancer Springer Journals

Improving the letters we write: an exploration of doctor–doctor communication in cancer care

Loading next page...
 
/lp/springer-journals/improving-the-letters-we-write-an-exploration-of-doctor-doctor-JpnjFZlVNd

References (16)

Publisher
Springer Journals
Copyright
Copyright © 1999 by The Author(s)
Subject
Biomedicine; Biomedicine, general; Cancer Research; Epidemiology; Molecular Medicine; Oncology; Drug Resistance
ISSN
0007-0920
eISSN
1532-1827
DOI
10.1038/sj.bjc.6690374
Publisher site
See Article on Publisher Site

Abstract

British Journal of Cancer (1999) 80(3/4), 427–437 © 1999 Cancer Research Campaign Article no. bjoc.1998.0374 Improving the letters we write: an exploration of doctorÐdoctor communication in cancer care 1 1,2 3 D McConnell , PN Butow and MHN Tattersall 1 2 3 Medical Psychology Unit and Department of Psychological Medicine, University of Sydney, NSW 2006, Australia; Department of Cancer Medicine, University of Sydney, NSW 2006, Australia Summary Referral and reply letters are common means by which doctors exchange information pertinent to patient care. Twenty-eight semi- structured interviews were conducted exploring the views of oncologists, referring surgeons and general practitioners. Twenty-seven categories of information in referral letters and 32 in reply letters after a consultation were defined. The letters to and from six medical oncologists relating to 20 consecutive new patients were copied, and their content analysed. Oncologists, surgeons and general practitioners Australia wide were surveyed using questionnaires developed on data obtained above. Only four of 27 categories of referral information appear regularly (in . 50%) in referral letters. Oncologists want most to receive information regarding the patient’s medical status, the involvement of other doctors, and any special considerations. Referring surgeons and family doctors identified delay in receiving the consultant’s reply letter as of greatest concern, and insufficient detail as relatively common problems. Reply letters include more information regarding patient history/background than the recipients would like. Referring surgeons and family doctors want information regarding the proposed treatment, expected outcomes, and any psychosocial concerns, yet these items are often omitted. Consultants and referring doctors need to review, and modify their letter writing practices. Keywords: letters; referrals; cancer consultation; multidisciplinary care; communication Optimal patient care hinges at least in part on adequate and timely wished to receive information regarding the prognosis and what the exchange of information between treating doctors (Newton et al, patient had been told, yet less than 20% of letters adequately 1992). The referral and reply letters are the most common means by covered these topics. The more recent study, conducted in which doctors exchange information pertinent to patient care Australia, examined 94 reply letters sent by one oncologist (Tattersall et al, 1995). If these letters meet the respective needs of (Tattersall et al, 1995). A questionnaire was sent to 55 GPs and 53 consultants and referring doctors, discontinuity in care, unnecessary referring specialists who had received a letter from the oncologist repetition of diagnostic tests and poor patient outcomes such as asking them to rate each of 14 items as essential, useful, of little anxiety, dissatisfaction and loss of confidence in medical practitioners use, or of no use. may be avoided (Cummins et al, 1980; McPhee et al, 1984; Hull and The majority of respondents (n 5 95) rated the following items Wosterman, 1986; Nutting et al, 1992; Graham, 1994; Epstein, 1995). as essential: diagnosis, clinical findings, test results, further tests, Few studies have investigated the information content of doctors’ treatment options and recommendations, prognosis, and likely letters and/or information preferences of doctor recipients. benefits and side-effects. Less than 50% of doctors regarded Only one study has examined referral letters in the cancer care details of the patients’ presenting history, drug or social history as setting. In this Australian study a limited audit was made of 103 essential. Content analysis of the reply letters found that they consecutive new patients seen by one radiation oncologist (Graham, usually did not specify prognosis, give recommendations of 1994). Of the 80 letters available, 95% reported the diagnosis, but further tests, or specify the likely side-effects of treatment, and only 56% provided a history of the current illness. Less than half the more commonly than referring doctors desired, included details on referrals detailed clinical findings or included information on past presenting history, past medical, drug and social history. The history, social history, medications and allergies. The author extent to which these findings can be generalized, however, is concluded that relevant and important information was not commu- unknown. The letters analysed were from only one oncologist and nicated in referral letters. criteria ‘presumed ideal’ were used for the content analysis, and to Only two studies have specifically investigated the content of identify doctors’ information preferences. letters from oncologists, and the information preferences of the We have conducted a comprehensive audit of referral and reply recipients. Bado and Williams (1984), in their survey of 73 general letters to and from Australian oncologists and explored their infor- practitioners (GPs), reported that technical topics, such as diag- mation preferences and those of referring doctors (surgeons and nosis, findings on investigation and treatment details, were more GPs). Our objectives were as follows: important than social topics. More than 80% of GPs, however, • to determine the purpose/function and preferred content of referral and reply letters as perceived by oncologists and refer- Received 29 June 1998 ring doctors respectively Revised 10 October 1998 • to obtain a representative view of oncologists concerns with Accepted 20 November 1998 referral letters and referring doctors concerns regarding reply Correspondence to: MHN Tattersall letters 427 428 D McConnell et al Table 1 What oncologists want in most/all cases and what they get in referral letters Content items Medical oncologists Radiation oncologists Actual content (n 5 113) (n 5 43) (n 5 89) % Factor 1 – Patient’s wishes/concerns Mean 5 2.58 Mean 5 3.02 % most/all % most/all How the patient is coping 48.2 63.4 6.7 The patient’s information, involvement and treatment preferences 51.8 64.3 18 Impact of the cancer and treatment on the patient’s work, 37.8 56.1 6.7 leisure and self-care activities Factor 2 – Patient’s background Mean 5 2.61 Mean 5 2.82 %% Family history of cancer 39.3 52.4 3.4 Social history – lifestyle e.g. smoking, drinking 43.2 50 10.1 Social history – employment and home situation 44.2 42.9 15.7 Clinical/findings on examination 57.5 88.1 15.7 Factor 3 – Patient’s medical status Mean 5 3.45 Mean 5 3.43 %% Inter-current medical conditions 85 81 22.5 Past medical history 68.2 69 20.2 Current medication 93.8 83.3 21.3 History of presenting problem 80.4 87.5 82 Factor 4 – Involvement of other doctors and their views Mean 5 3.33 Mean 5 3.67 %% Involvement of other doctors in the case 81.3 90.5 23.6 Referring doctor’s view of his/her continuing 69.9 90.5 19.1 involvement in the case What opinions have been expressed by 69.9 85.7 11.2 other doctors about patient management What the patient has been told 80.5 90.5 13.5 The referring doctor’s thoughts on what may 52.3 73.8 32.6 be appropriate management Any factors possibly mitigating against particular 85 100 5.6 treatments or treatment arrangements Tests/findings on investigation 98.2 100 61.8 Factor 5 – Special considerations Mean 5 3.25 Mean 5 3.31 %% Concerns about psychiatric/social problems 75.9 78 3.4 Concerns about patient compliance 68.8 76.2 2.2 Concerns about patient understanding 67.9 73.8 2.2 Wishes/concerns of the patient’s family 58 75.6 1.1 Need for an interpreter 87.4 78.6 1.1 Information regarding any formal clinical trials the 71.4 69 1.1 patient is on or is eligible for Additional items % Mean % Mean Clearly stated reason for referral 98.2 3.94 97.6 3.9 78.7% Provisional diagnosis 88.3 3.51 97.6 3.9 88.8 Copies of test results/reports/films 94.6 3.85 95.2 3.88 N/A Items are listed in order of factor loading. Discrepancies in which . 75% of both medical & radiation oncologists want an item in most/all cases, but , 25% of a b letters actually contain this item are shown in bold. Denotes a significant difference between mean scores at the level of P , 0.05. Denotes a significant difference between mean scores at the level of P , 0.01. • to determine what information is ‘typically’ contained in 28 semi-structured interviews with doctors were conducted referral letters to oncologists, and their reply letters including seven with oncologists from three Sydney hospitals, ten • to prepare a template of referral and reply letters which may with surgeons and 11 with GPs practising in the Sydney enhance communication between referring doctors and oncolo- Metropolitan area. Two interviews were conducted by telephone gists. with GPs in rural areas. All other interviews were conducted in person. The interviews explored doctors’ views on referral communications with a focus on their information needs. All inter- METHOD views were audiotaped and transcribed. The interview data were analysed using the constant-compara- Stage 1 – qualitative phase tive method proposed by Glaser and Strauss (1967). Put simply, In Stage 1, three medical and three radiation oncologists were this involves coding each unit of meaning (i.e. specific response), invited to participate in an interview and to provide contact details and comparing and contrasting these to identify recurring regular- of their last four new patients, their referring doctors and GPs. An ities and discrete categories. This resulted in the development of invitation to participate was then sent to these doctors. A total of an analytic framework of 27 discrete categories of information for British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 429 Table 2 What surgeons and GPs want in most/all cases and what they get in reply letters Content items Surgeons GPs Actual content (n 5 99) % Factor 1 – History/background Mean 5 2.53 Mean 5 3.03 % most/all cases % most/all cases Lifestyle risk factors 35.6 56.1 68.7 Family history of cancer 40.7 70.4 66.7 History of presenting problem 42.6 73.6 97 Past medical history 24.6 37.9 82.8 Social history 24.1 50.4 69.7 Current medication 55.6 89.8 73.7 Inter-current medical conditions 59.2 69.4 64.6 Restatement of reason for referral 31.5 75.7 6.1 Factor 2 – Psycho-social concerns Mean 5 3.13 Mean 5 3.61 %% Concerns about patient understanding 64.9 84.3 6.1 Concerns about psychiatric/social problems 59.4 83.1 1 Concerns about patient compliance 68 88.8 2 Patient’s wishes/expectations regarding information disclosure, 66.7 86.1 26.3 decision making/treatment (3) Impact of cancer and/or treatment on patient’s work, 53.7 79.6 6.1 leisure and self-care activities Likely prognosis (5) 81.5 95.4 31.3 How patient is coping/feeling about diagnosis/prognosis/treatment 68.5 87.9 16.2 Factor 3 – Examination and investigation findings Mean 5 3.66 Mean 5 3.92 %% Tests/findings on investigation 92.6 98.1 41.4 Clinical/findings on examination 74.1 95.3 89.9 Treatment recommendation 94.4 100 85.9 Diagnosis/provisional diagnosis 86.8 100 96 Factor 4 – Future management/expectations Mean 5 2.90 Mean 5 3.71 %% Likely short- and long-term side-effects 58.4 93.4 16.2 Suggestions for management of side-effects 43.6 91.5 5.1 Indicators for unscheduled review by the oncologist 52.8 85.8 8.1 Aim of treatment e.g. curative or palliative (5) 81.1 97.2 40.4 Intention of the oncologist to contact the referring Dr/GP in 54.7 87.8 51.5 the future (4) Factor 5 – Treatment/management plan Mean 5 3.57 Mean 5 3.78 %% The oncologist’s follow-up plan 90.5 98.1 67.7 Involvement of other doctors in the case 80.8 89.7 32.3 Rationale for recommended treatment (3) 79.2 91.6 66.7 Arrangements made for treatment, i.e. where and when 77.4 85.9 33.3 What the patient has been told 88.4 92.4 49.5 Anything specific the oncologist would like the referring Dr/GP 92.4 99.1 14.1 to do. Treatment options 84.9 94.4 31.3 Additional item % Mean % Mean Information regarding any formal clinical trial discussed with 75.5 3.30 85 3.55 10.1 the patient ( ) Indicates that the item also loaded on the factor shown in brackets. Items are listed in order of factor loading. Percentage figures shown in bold highlight a b discrepancies between actual content and preferences of . 50%. 5 P , 0.01; 5 P , 0.001. referral letters and 32 for consultation reply letters (Tables 1 and patients’ medical files were then traced, and referral and reply 2). Common problems encountered in communication between letters photocopied. During data collection, 21 files were not avail- doctors were identified. This analytic framework was used in able and an additional ten referral letters were absent from files. A Stage 2 to analyse the content of referral and reply letters and sample of 89 referral letters and 99 consultation reply letters was provided the basis for the development of questionnaires used in therefore obtained. Stage 3 to survey each group of doctors. Most of the referral letters (77%) were from surgeons or other medical specialists, and 93% were outpatient referrals. The content of each letter using the analytic framework developed in Stage 1 Stage 2 – Content analysis of referral and reply letters was determined by simply noting whether each item of informa- Six medical oncologists from two Sydney hospitals were asked tion was present. The first and third author each analysed a random to provide a list of their last 20 consecutive new patients. The selection of ten letters. Agreement between raters was moderately © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 430 D McConnell et al Table 3 Sample characteristics: Stage 3 to the patient’s situation, and the reason for referral, (b) contributes to assessment by reducing the likelihood of relevant information Characteristics Surgeons GPs Oncologists being overlooked, (c) improves efficiency and quality of care by reducing unnecessary duplication of tests, and providing a focus Sample size n 5 55 n 5 108 n 5 156 Gender for history taking, and (d) provides the groundwork for ongoing Male 54 (98%) 65 (60%) 133 (85%) care and communication. Oncologists reported that missing Female 1 (2%) 43 (40%) 23 (15%) reports or tests results and insufficient detail in the referral letter Years of experience were the most frequent concerns and these were more problematic Mean 19.74 16.59 12.56 than any other (P 5, 0.05). Range 4–40 2–50 0–39 Speciality N/A N/A General surgeon 32 (57%) Actual vs preferred content of referral letters Other surgeon 23 (43%) Twenty-seven categories of information sought in referral letters Average number of cancer Data not Data not were identified in Stage 1. The questionnaire explored oncolo- patients per year collected 2 (2%) collected , 1 33 (31%) gists’ preferences for information in new patients referral letters, 1–5 27 (25%) and respondents indicated on a four-point scale the proportion of 6–10 45 (42%) cases (none, some, most, or all) in which they would like to . 10 1 receive each of the 27 items of information. The aim was to iden- tify ‘in-general’ preferences and priorities for information and to examine current practice in light of these. high at 86%, supporting the reliability and utility of the informa- To identify groups/clusters of items, a factor analysis was tion categories. Upon completion of the coding, nine randomly undertaken. With oblique rotation, a five-factor resolution selected referral and reply letters were recoded to examine intra- emerged, accounting for 51.7% of the variance. Two items, rater reliability. A high level of intra-rater agreement was obtained ‘reason for referral’ and ‘provisional diagnosis’ did not load on at 98%. any of the factors above 0.325 and were therefore considered sepa- rately in subsequent analyses. Table 1 shows the distribution of the 25 items composing the five factors, the percentage of medical and Stage 3 – survey radiation oncologists wanting each item in most/all cases, and the In Stage 3, questionnaires for oncologists, surgeons and GPs were proportion of letters analysed in Stage 2 in which each item was developed based on data obtained in Stage 1 (Appendix 1). present. Oncologists were asked to indicate (a) their preferences for 27 It is evident that a discrepancy exists between information items of information in a referral letter, (b) the frequency with contained and information desired in referral letters. Only four out which they encountered seven common difficulties in referral of 27 items appear regularly (i.e. in more than 50%) of referral communications and (c) if and when a telephone call was preferred letters, namely, the provisional diagnosis, history of the presenting to a letter. Mirroring this, surgeons and GPs were asked (a) their problem, clearly stated reason for referral and findings on investi- preferences for 32 items of information in letters of reply, (b) the gation. On these four items only, referral letters appear to meet frequency of five common problems in reply letters, and (c) when a oncologists’ information needs/preferences. telephone call is preferred to a letter. The questionnaires were Seven items of information wanted by more than 75% of piloted with three oncologists, surgeons and GPs to ensure clarity medical and radiation oncologists in most or all cases were docu- in wording and format. The resulting questionnaire was sent to all mented in less than 25% of letters. Specifically these items are: members of the Medical Oncology Group of the Royal (1) inter-current medical conditions, (2) current medication, (3) Australasian College of Physicians (n 5 148), and all surgeons (n involvement of other doctors in the patient’s care, (4) what the 5 84) and radiation oncologists (n 5 56) who are members of the patient has been told, (5) any factors possibly mitigating against Clinical Oncological Society of Australia (COSA). The sample of particular treatments, (6) concerns about psychiatric/social prob- 200 GPs was drawn from the Directory of Members of the Royal lems, and (7) need for an interpreter. Australian College of General Practitioners which lists almost 10 In interviews and surveys, oncologists identified circumstances 000 members. The sample of GPs was selected using a randomized in which their information needs/preferences may very. Several block design to ensure a representative proportion from each State variables relating to individual patient characteristics and the and Territory. In total, 113 medical oncologists, 43 radiation oncol- nature of the referral were identified. These variables include: (1) ogists, 55 surgeons and 108 GPs returned completed questionnaires whether the patient is an in-patient or out-patient, (2) whether the representing a 76%, 77%, 65% and 54% response rate respectively. doctors interact in a multi-disciplinary clinic, (3) whether the It was not possible to establish the existence of bias introduced by patient is referred preoperatively or post-operatively, (4) whether these response rates which were rather low in the latter two groups. the cancer problem is simple or complex, (5) how well the refer- Some demographic details are presented in Table 3. ring doctor and oncologist know each other, and (6) whether there are significant psycho-social concerns about the patient. Examining how these variables may affect information needs was RESULTS beyond the scope of this study, and they are not allowed for in the presentation of preferences which follows. The referral letter – views of oncologists Analysis of interview data and responses to the survey question Perceived problems with referral letters concerning the function of the referral letter identified four Oncologists interviewed in Stage 1 identified seven concerns common themes. The letter (a) provides background information with referral letters (Table 4). In the questionnaire, we asked British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 431 Table 4 Perceived frequency in which each problem with referral letters occurs (n 5 156) Problems – in descending order from most to Mean [95% CI] [1 5 always, 7 5 never] least frequent Missing reports/test results – i.e. pathology, 3.13 (2.94–3.32) X-ray films, operation report Insufficient information and detail in the 3.46 (3.26–3.66) referral letter No referral letter received prior to or at the time 4.05 (3.82–4.27) of the consultation Hand-written referral letters which are difficult 4.19 (3.98–4.41) or impossible to read Unclearly specified reason for referral 4.89 (4.64–5.14) No referral letter received at all 5.02 (4.78–5.26) Unnecessary information in the referral letter 5.70 (5.50–5.89) Table 5 Perceived frequency of problems with reply letters Mean (95% CI) [1 5 Always, 7 5 Never] Perceived problems Surgeons GPs 1. Reply letters arriving late – not promptly 4.3774 (3.934–4.821) 3.6944 (3.412–3.977) 2. Unnecessary information in the reply letter 4.7170 (4.293–5.140) 5.9907 (5.729–6.252) 3. Insufficient information in the reply letter 5.3396 (4.985–5.694) 4.5648 (3.460–5.669) 4. No reply letter received at all 5.6226 (5.210–6.035) 4.6262 (4.328–4.925) 5. Letters that are too technical and 6.1887 (5.887–6.490) 5.8333 (5.577–6.090) consequently difficult to comprehend oncologists to indicate on a seven-point scale the frequency with consultation, (2) when there is sensitive information to convey, which each of these seven problems occur (from always to never), e.g. if the patient is dissatisfied with other doctors or their manage- and then to identify and rank the three that are most problematic. ment to date, (3) if there are personality or psychological issues Mean scores with 95% confidence intervals were computed. that may affect compliance with treatment recommendations, and Oncologists perceive that missing reports or test results and insuf- (4) if the problem is complex and difficult to relate in a letter and ficient detail in the referral letter occur significantly more often multiple opinions have been sought. than any other problem. These concerns were perceived to be significantly more problematic than any other (P , 0.05). The reply letter – views of referring surgeons and GPs Comparison of medical and radiation oncologists Actual vs preferred content of post-consultation reply letters Figures in Table 1 suggest that radiation oncologists want more Thirty-two categories of information were identified in Stage 1 as information than medical oncologists do in most categories. To components of post-consultation reply letters from oncologists. In statistically explore this finding, the mean score of items in each Stage 2, the actual content of the sample of post-consultation reply factor (where 1 5 in no cases and 4 5 in all cases) were computed letters from radiation and medical oncologists were analysed, and separately for each specialty group and compared using t-tests for in Stage 3, preferences of surgeons and GPs for these items of independent samples. Radiation oncologists on average want more information were sought. Surgeons and GPs indicated on a four- information than medical oncologists concerning patients’ point scale the proportion of cases (none, some, most, or all) in wishes/concerns (P , 0.05) and the involvement of other doctors which they liked to receive letters covering each of the 32 items of in the case (P , 0.01). information identified in Stage 1. Our aim was to identify ‘in- Both medical and radiation oncologists primarily want informa- general’ preferences and priorities for letter content, and then to tion regarding the patient’s medical status, the involvement of evaluate a sample of reply letters with reference to these. other doctors and special considerations. Information concerning To identify groups of related items, a factor analysis was the patient’s wishes/concerns and the patient’s history/background conducted using the data from the survey of referring surgeons and appear to be of secondary importance. GPs. With varimax rotation, a five-factor resolution was obtained accounting for 48.4% of the variance. One item failed to load on When would oncologists like the referring doctor to phone any factor above 0.325 and was therefore analysed separately, them? namely, ‘information regarding any formal clinical trial discussed Most oncologists (73%) indicated that they would like the refer- with the patient’. The five groups, and items loading are shown in ring doctor to phone them (1) when the patient needs an urgent Table 2. Also shown is the percentage of surgeons and GPs © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 432 D McConnell et al Table 6 Information prompt sheets Referral letters ® © Reason for referral ® © Provisional diagnosis ® © Succinct history of the problem ® © Relevant information on patient’s medical status – current medications, inter-current medical conditions and relevant past medical history ® Clinical/findings on examination ® © Information on tests performed and results ® Patient’s wishes and concerns, e.g. how the patient is coping, and their information, involvement and treatment preferences ® © What the patient has been told ® © Involvement of other doctors; what role the referring doctor expects to play; other opinions on management ® © Any factors possibly mitigating against particular treatments or treatment arrangements ® © Special considerations, e.g. psychiatric/social problems, concerns regarding compliance or patient understanding, need for an interpreter, and any concerns/wishes of patient’s family ® © Copies of relevant test results/reports Reply Letters . Restatement of reason for referral . History of presenting problem, family history of cancer, current medication, intercurrent medical conditions ! . Clinical findings on examination; tests/findings on investigation ! . Diagnosis and likely prognosis ! . Treatment options, treatment recommendation with rationale, treatment aim ! . Patient’s wishes and expectations, and how he/she is coping ! . Psycho-social concerns, e.g. patient understanding, psychiatric/social problems ! . Management plan – arrangements, follow-up, and involvement of other doctors . Likely short- and long-term side-effects, and suggestions for the management of these ! . What the patient has been told ! . How and when to contact the oncologist/consultant ® 5 Radiation oncologist, © 5 Medical oncologist, ! 5 Surgeon, . 5 GP. wanting each items in most or all cases, and the percentage of Perceived problems with reply letters reply letters including each item. Five potential problems with reply letters were identified in Stage These data suggest that oncologists’ letters do not provide all 1 interviews (see Table 4). The surgeons and GPs surveyed indi- the information surgeons and GPs want. Oncologists’ letters cated on a seven-point scale how often they perceive that each commonly provide details on examination and investigation find- problem occurs, and identified and ranked the three that are most ings (factor 3), and these items are those most often desired by problematic. Mean scores and 95% confidence intervals were surgeons and GPs. However, the majority of surgeons and GPs computed for each identified problem. Both surgeons and GPs want details of the treatment/management plan (factor 5), future perceive that delay in receiving the reply letter is the most management/expectations (factor 4), and psycho-social concerns frequently occurring problem, and the problem which is of most (factor 2), yet these items are rarely mentioned in letters. concern to them. Superfluous information in the reply letter is Oncologists’ letters also frequently detail the patient’s back- perceived by surgeons to be the next most common problem. GPs’ ground/history (factor 1), which make up six of the ten most however, perceive this to be the least common problem. common items in reply letters. These items, however, are those The preferences of surgeons and GPs for information least often desired by referring surgeons and GPs. The data in Table 2 suggest that the information needs/preferences Several circumstances influencing referring doctor’s informa- of referring surgeons and GPs differ. To test this observation, the tion preferences were identified. These include: (1) how well the mean score of items in each factor (where 1 5 in no cases, and 4 5 referring doctor knows the oncologist, (2) whether there are in all cases) for surgeons and GPs were computed and compared routine clinical meetings between the referring doctor and oncolo- using t-tests for independent samples. The results indicate that GPs gist, (3) the reason for referral – e.g. for second opinion or to take on average want more information than surgeons in every cate- over patient management, (4) whether the patient consultation is gory. Both surgeons and GPs place highest priority on receiving pre- or post-surgery, (5) whether the patient is an in-patient or out- details of the examination and investigation findings, and the patient, (6) whether the cancer is rare or common, and (7) whether proposed treatment/management plan. the treatment recommended is standard or not. However, exam- ining how these variables may affect information needs/prefer- When would surgeons and GPs like the oncologist to phone ences of referring specialists and GPs was beyond the scope of this them? study and they are not allowed for in the presentation of prefer- Sixty per cent of surgeons and 78% of GPs identified circum- ences which follows. stances in which a reply letter is insufficient and a phone call British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 433 desirable. Specific circumstances in which referring doctors would has been told, the treatment options, aim of treatment and likely like the oncologist to phone them are (1) when urgent issues arise, prognosis, the involvement of other doctors in the case, and (2) when the treatment proposed is unconventional, and (3) when anything specific the oncologist would like the referring doctor/GP the oncologist is uncertain about the preferred management. A to do. Previous studies have also identified the absence of informa- telephone call is also favoured when (4) divergent views exist on tion on prognosis and what the patient has been told as significant treatment approach and (5) if the treatment recommendation is gaps in the information content of ‘typical’ reply letters. different to that which the referring doctor thought appropriate at It is common practice for oncologists to send GPs a copy of the the time of referral. reply letter to the referring surgeon without alteration. Previous studies have either looked at the information needs of GPs alone, or grouped them together with referring specialists. This study DISCUSSION compared the information preferences of surgeons and GPs, and Doctors write many referral letters either to clinical colleagues or our findings suggest that one reply letter may not adequately meet to diagnostic service providers. Specialist physicians write letters the needs of both. Information preferences appear to be the same, in reply to referring doctors after new patient consultations or with both surgeons and GPs wanting information concerning follow-up visits, and to clinicians caring for patients at home examination and investigation findings most, and information following discharge from hospital. Previous studies suggest the regarding patient history/background least. However, the results of content, legibility, speed of receipt and relevance of doctors’ this study indicate that GPs want significantly more information letters are often deficient and/or do not meet expectations. We than surgeons in every category. These results may explain the have conducted an information audit of referral and reply letters, differences between surgeons and GPs in their perceptions of prob- interviewed and surveyed a sample of referring doctors and oncol- lems with reply letters. Superfluous information is perceived by ogists concerning their preferences and experience with doctors’ surgeons to be the second most common problem with reply letters. The results of this study suggest the need for doctors to letters. GPs, however, perceive this to be the least frequently review, and modify their letter writing practices. occurring problem, if in fact a problem at all. We found that referral letters typically include a statement of the reason for referral, some history of the problem, a provisional Implications for practice diagnosis and description of the findings on investigation. Whilst these items are among the ‘most wanted’, oncologists in this study The findings of this study raise doubts as to whether referral and have clearly articulated a ‘wish’ for a range of additional items of reply letters fulfill their perceived functions. Modifying letter information. At the top of oncologists’ ‘wish list’ is information writing practices may be a relatively simple and effective means concerning the patient’s medical status, the involvement of other of improving doctor–doctor communication and hence, patient doctors and any special considerations. Many oncologists also understanding and outcomes. Referring doctors could improve prefer letters that outline the patient’s history and their wishes and communication between themselves and medical and radiation concerns, but this information appears to be of secondary impor- oncologists by ensuring that available test results/reports accom- tance presumably because these items would be sought during pany the referral letter, by mailing the referral letter to the oncolo- history taking. Radiation oncologists appear to want more infor- gist prior to the consultation and giving a copy to the patient. An mation in referral letters than medical oncologists, particularly in information prompt sheet for referral letters and letters of reply is the areas of patients’ wishes/concerns and the involvement of provided in Table 6. other doctors. Given the significant discrepancy between informa- Medical and radiation oncologists could take several steps to tion desired and information contained, it is not surprising that improve communication with referring surgeons and GPs. Letters oncologists perceive that insufficient information and detail is one should be sent soon after the consultation, since delay in receiving of the two most frequently occurring problems with referral letters. the reply letter is a major concern of both surgeons and GPs. Post-consultation reply letters from oncologists are not meeting Oncologists’ letters should not recount all aspects of the patient the information preferences of referring surgeons and GPs. From history. However, these letters should document the results of the letter writer’s perspective, the reply letter also functions as examination and investigations, the treatment options and a consultation record. Kamien (1995) has highlighted this proposed management plan, state the prognosis and what the dichotomy of purpose, and argued that it must be resolved in the patient has been told, and outline any psycho-social concerns. interests of good communication. Should we write two letters, one Although a case can be made for writing two letters, one for a that is filed in the notes as a record of the consultation, and the referring surgeon (if relevant) that is short and succinct, and one second that is prepared specifically to inform the referring doctor for GPs that is more comprehensive, this is clearly not practical. and meet their information needs? For GPs’ standard information sheets may be included with the Our results confirm previous findings. Tattersall et al (1995) reply letter concerning the cancer type, potential side-effects of the concluded that reply letters, more often than is desired, contain treatment proposed and recommendations for their management. information concerning patient history. This study confirmed that More than 90% of GPs want this information and less than 20% of items of information concerning patient history/background are oncologist reply letters currently provide any of these details. among the most common items in reply letters, but are the least desired. Surgeons and GPs prefer details concerning the treat- Future research ment/management plan, future management/expectations and psycho-social concerns, yet these are rarely provided in reply Future research should examine how the information needs/prefer- letters. There were several items desired by surgeons and GPs in ences of oncologists and referring doctors may vary with the more than 80% of cases, but included in less than 50% of letters. circumstances identified in this study. Such research will permit These were findings on examination, details of what the patient doctors to better predict and tailor their letters to referring and © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 434 D McConnell et al Epstein RM (1995) Communication between primary care physicians and other doctors. In addition, referral and reply letters, which incorpo- consultants. Arch Family Med 4: 403–409 rate the recommendations of this study, should be evaluated to Glaser BG and Strauss AL (1967) The Discovery of Grounded Theory. Strategies for determine whether they result in increased satisfaction on the part Qualitative Research. Aldine: New York of recipients, whether they fulfill their perceived functions as iden- Graham PH (1994) Improving communication with specialists. The case of an tified in this study and whether they result in better patient oncology clinic. Med J Aust 160: 625–627 Hull FM and Westerman RF (1986) Referral to medical outpatients department at outcomes. teaching hospitals in Birmingham and Amsterdam. Br Med J 293: 311–314 Kamien M (1995) Writing to referring doctors. Aust NZ J Med 25: 463–464 ACKNOWLEDGEMENTS McPhee SJ, Lo B, Saika GY and Meltzer R (1984) How good is communication between primary care physicians and subspecialty consultants? Arch Int Med This study is supported by a NSW Cancer Council Patient Care 144: 1265–1268 Award. Newton J, Eccles M and Hutchinson A (1992) Communication between general practitioners and consultants: what should their letters contain? Br Med J 304: REFERENCES 821–824 Nutting PA, Franks P and Clancy CM (1992) Referral and consultation in primary Bado W and Williams CJ (1984) Usefulness of letters from hospitals to general care: do we understand what we are doing? J Family Pract 35: 21–23 practitioners. Br Med J 288: 1813–1814 Tattersall MHN, Griffin A, Dunn SM, Monaghan H, Scatchard K and Butow PN Cummins RO, Smith RW and Inui TS (1980) Communication failure in primary (1995) Writing to referring doctors after a new patient consultation. What is care. Failure of consultants to provide follow-up information. JAMA 243: wanted and what was contained in letters from one medical oncologist? Aust 1650–1652 NZ J Med 25: 479–482 APPENDIX 1 The initial patient consultation with a medical or radiation oncologist: doctor and patient information preferences QUESTIONNAIRE FOR ONCOLOGISTS What information do onocologists want to receive with a new patient referral? What concerns to oncologists have about referral letters? What are the views of oncologists about providing patients with a post-consultation letter? This questionnaire is primarily concerned with these three questions. Please take the time (approximately 15 minutes) to fill it in. Your views are important in order to obtain a representative view of oncologists. If you have any questions about this project, please contact Mr David McConnell on (02) 9515 8160. Your answers will remain strictly confidential. Thank you in advance for your participation. Part A – Treatment decision making and working with other doctors 1) In your opinion, how should treatment decisions be made? Please tick the statement which best describes your opinion (please tick one box only) n n The doctor should make the decisions based on what he/she determines to be the best treatment for the cancer n n The doctor should make the decisions but consider the patient’s priorities and quality of life n n The patient and the doctor should make the decisions together n n The patient should make the decisions, but consider the doctor’s opinion n n The patient should make the decisions using all they know or learn about their treatment options 2) Please indicate the extent to which you agree/disagree with each of the following statements by circling the number which best repre- sents your view (1 5 strongly agree, 7 5 strongly disagree) Generally speaking, for patients who may see other doctors… Strongly Strongly Agree Disagree A. Oncologists should try to ensure the information 1 2 3 4 5 6 7 they give to patients is compatible with that likely to be given by other doctors B. Oncologists should consider the views of the 1 2 3 4 5 6 7 patient’s GP in determining the treatment plan C. Oncologists should consider the views of doctors 1 2 3 4 5 6 7 from other specialities in determining the treatment plan D. Oncologists should share follow-up with doctors 1 2 3 4 5 6 7 from other specialities E. Oncologists should share follow-up with the patient’s GP 1 2 3 4 5 6 7 F. A patient’s cancer care should be jointly managed 1 2 3 4 5 6 7 by the oncologist and the GP G. A patient’s cancer care should be jointly 1 2 3 4 5 6 7 managed by the oncologist and doctors from other specialities H. A patient should be referred to an oncologist prior to surgery 1 2 3 4 5 6 7 British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 435 Part B – Information accompanying referrals 3) In what proportion of cases would you like to receive each item of information listed below in a referral letter? If you tick most or some for any item of information, please specify the circumstances in which you want that information from the referring doctor. Items of information With all With With With no Please use this space to specify referrals most some referrals referrals referrals (specify) (specify) 1. Patient’s social history, n n n n n n n n e.g. employment, home situation 2. Reason for referral n n n n n n n n 3. History of presenting problem n n n n n n n n 4. Family history of cancer n n n n n n n n 5. Social history – lifestyle, n n n n n n n n e.g. smoking, drinking 6. Past medical history – unrelated to n n n n n n n n the presenting problem 7. Inter current medical conditions – physical n n n n n n n n & psychiatric 8. Current medication n n n n n n n n 9. Clinical findings: results of physical n n n n n n n n examination 10. What tests have been done or n n n n n n n n arranged by the referring doctor & a summary of the main findings 11. Diagnosis/provisional diagnosis n n n n n n n n 12. Referring doctor’s thoughts on what n n n n n n n n may be appropriate management 13. What other opinions have been n n n n n n n n expressed by other doctors about patient management 14. Any factors possibly mitigating n n n n n n n n against certain treatments or treatment arrangements – medical, psycho-social, or demographic 15. Referring doctor’s view of his/her n n n n n n n n continuing involvement in the case 16. Involvement of other doctors in the case n n n n n n n n 17. What the patient has been told regarding n n n n n n n n diagnosis, prognosis, treatment options 18. The patient’s wishes, expectations or n n n n n n n n concerns regarding information disclosure, decision making, treatment 19. How the patient is coping and/or feeling n n n n n n n n about their diagnosis, prognosis or treatment 20. Impact of the cancer & its treatment on n n n n n n n n the patient’s work, leisure and self care activities 21. Any concerns about how much the n n n n n n n n patient understands 22. Any concerns about psychiatric and/or n n n n n n n n social problems 23. Any concerns about patient compliance n n n n n n n n willingness to accept advice © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437 436 D McConnell et al Items of information With all With With With no Please use this space to specify referrals most some referrals referrals referrals (specify) (specify) 24. Whether an interpreter is required for the n n n n n n n n consultation [if the patient has difficulty speaking English] 25. Information regarding any formal n n n n n n n n clinical trials the patient is on, has been offered, or is eligible for 26. Any wishes/concerns of the patient’s n n n n n n n n family, e.g. about the disclosure of information to the patient 27. Copies of test results, n n n n n n n n e.g. pathology report, X-ray films Would you like to receive any other information from the referring doctor? If so, please specify on the back of this page. 4) Is there any information you would prefer to receive over the phone, or circumstances in which you would like the referring doctor to phone you? n n Yes n n No If yes, please specify. 5) How is the information you receive from the referring doctor helpful? What purpose does it serve? 6) i. You may have experienced the following problems with referral letters. Please circle the number which best represents how often each occurs. (1 5 always, 7 5 never). Always Never A. Missing reports/test results – i.e. pathology, 1 2 3 4 5 6 7 X-ray films, operation report etc. B. Hand-written referral letters which are difficult 1 2 3 4 5 6 7 or impossible to read C. Unclearly specified reason for referral 1 2 3 4 5 6 7 D. Insufficient information and detail in the referral letter 1 2 3 4 5 6 7 E. Unnecessary information in the referral letter 1 2 3 4 5 6 7 F. No referral letter received prior to or at the time of consultation 1 2 3 4 5 6 7 G. No referral letter received at all 1 2 3 4 5 6 7 Please list any other concerns you may have and indicate how often each occurs 12 3 4 5 6 7 12 3 4 5 6 7 7) ii. From the list above, which 3 concerns about referral letters are most problematic? Please list and rank these with 1 being the most problematic. 8) When would you ideally like to receive the referral letter? n n Prior to the patient consultation n n At the time of the patient consultation n n It doesn’t matter 9) In what format would you prefer the referral letter to be written? n n In narrative format n n In point form n n It doesn’t matter British Journal of Cancer (1999) 80(3/4), 427–437 © Cancer Research Campaign 1999 Doctor–doctor communication in cancer care 437 10) When a patient is not referred by their GP, how often do you practice each of the following activities? (1 5 always, 7 5 never) Always Never i. Send the GP a copy of the letter written to 1 2 3 4 5 6 7 the referring doctor ii. Write an additional letter to the GP 1 2 3 4 5 6 7 iii. Send the GP a copy of the letter written to 1 2 3 4 5 6 7 the referring doctor – with an additional post-script iv. Send the GP a copy of the letter addressed 1 2 3 4 5 6 7 to the referring doctor, but written with the GP in mind Part C – Patient information Several studies suggest that patients have difficulty remembering information conveyed in their initial consultation. We would like to obtain your views on 3 strategies which may address this problem. 11) A. Do you think patients should be offered a copy of the letter written to the referring doctor? n n Yes n n No n n It depends Please explain: B. Do you think patients should be offered an individualized/personal letter as a follow-up to their consultation with you? n n Yes n n No n n It depends Please explain: C. Do you think patients should be offered an audiotaped recording of their consultation with you? n n Yes n n No n n It depends Please explain: 12) Which of the above strategies for providing information do you most prefer? n n a copy of the letter written to the referring doctor n n an individualized/personal letter following their consultation n n an audiotaped recording of their consultation n n None of the above 13) In your opinion, are there any ‘better’ strategies (better than those listed above) to ensure that patients are adequately informed? n n Yes n n No If yes, please specify: 14) In what proportion of cases do you practice each of the following activities? Please tick. In all In most In some In no cases cases cases cases i. Dictate your letter to the referring doctor in front of the patient n n n n n n n n ii. Offer patients a copy of the letter written to the referring doctor n n n n n n n n iii. Offer patients an individualized/personal n n n n n n n n letter after the consultation iv. Offer patients an audiotaped recording of the consultation n n n n n n n n v. Offer patients general information booklets n n n n n n n n 15) Would you like to make any further comments about any of the issues raised in this questionnaire? Personal Details: 16) Your sex n n Male n n Female 17) Your speciality n n Medical Oncology n n Radiation Oncology 18) How would you best describe your current position? n n University appointment n n Visiting Medical Officer n n Staff specialist n n Private practitioner n n Other 19) In what institution is your main practice? n n Private hospital n n Teaching hospital n n District hospital n n Other 20) For how many years have you been a practising oncologist? years. © Cancer Research Campaign 1999 British Journal of Cancer (1999) 80(3/4), 427–437

Journal

British Journal of CancerSpringer Journals

Published: Apr 9, 1999

There are no references for this article.