You are here

What surgeons should tell patients with oesophago-gastric cancer: A cross sectional study of information needs

European Journal of Surgical Oncology (EJSO), 11, 39, pages 1278 - 1286



Evidence shows that patients with cancer have many information needs, but specific requirements of patients undergoing surgery for oesophago-gastric (OG) cancer have not been well explored. This study surveyed information needs of patients with OG cancer and explored associations between patient characteristics and information needs.


A validated questionnaire assessing the importance of information about i) the nature, process and prognosis of the cancer, ii) tests, iii) treatments, and iv) physical and psychosocial outcomes (quality of life information) was completed by patients before and after OG cancer treatment. Items responses were transformed to a 0–100 score and linear regression explored associations between clinical and socio-demographic patient characteristics and patient information needs.


Questionnaires from 136 respondents were received (response rate 60%, 25% pre-operative, 77% men, mean age 66). Most types of information were rated as important, with information about prognosis and quality of life issues being scored as highly important by over 112 (82%) patients. Linear regression showed that women rated information relating to treatment and psychosocial effects as more important than men (p < 0.038), but no other associations were identified.


Most patients with OG cancer want detailed information, especially information about prognosis and quality of life. It is recommended that surgeons provide this for patients alongside information that the surgeon considers important.

Keywords: Oesophagogastric cancer, Information needs, Consent.


Consultations between surgeons and patients before undergoing surgery are a routine part of surgical practice. These encounters aim to provide patients with information about their illness and treatment, however, studies show that there is often dissatisfaction with the amount or quality of information received and that clinicians underestimate the complexity of information desired by patients.1, 2, 3, 4, 5, 6, and 7 This has important implications because poor communication is associated with worse adjustment to illness, worse quality of life, professional and patient dissatisfaction with the consultation and increased litigation.8, 9, 10, 11, and 12 In addition, without appropriate information patients cannot contribute to the process of shared decision-making, as is recommended by several national organisations,4, 13, and 14 or make a legal or ethical authorisation for treatment to take place through informed consent.

Much research has therefore focused on understanding patients' information needs with a view to improving information provision. A systematic review in 2005 identified 112 studies that investigated cancer patients' desire for information 15 and documented 10 major topics that patients wished to discuss with doctors including treatment related details and prognosis. This review also highlighted that patients' characteristics influence their desire for information; with young, educated women of higher social class seeking more information than other groups of patients. The majority of these studies, however, focused on breast or a heterogeneous mix of cancer sites and there is a paucity of research on the needs of patients undergoing complex, high risk cancer surgery. Surgery for oesophagogastric (OG) cancer, for example, is one of the highest risk elective operations undertaken. Often it represents the only chance of long-term survival but this benefit is balanced against significant risks of complications and a potentially irreversible deterioration in quality of life and death.16 and 17 Patients with OG cancer therefore face enormous life-changing events and provision of high quality information and support is an important part of patient care.

The aim of this study is to assess OG cancer patients' desire for information and to examine associations between patients' characteristics and information needs. It forms phase one of a broader research programme that will define a baseline information, or “core disclosure”, set that could be communicated to all patients undergoing OG cancer surgery to improve information provision.


Patients and data collection

Consecutive patients were identified from prospective multi-disciplinary cancer team records of the United Kingdom (UK) Avon, Somerset and Wiltshire Cancer Network between January 2005 and October 2008. Patients who had completed, or were undergoing curative treatment including surgery alone, neoadjuvant chemotherapy and surgery, or definitive chemoradiotherapy for adenocarcinoma of the oesophagus or stomach or oesophageal squamous cell cancer were included. Excluded were patients selected for palliative treatments, those with evidence of concurrent malignancy and patients who could not speak or understand English.

Potentially eligible patients were sent an introductory cover letter and information by post and were asked to return a reply slip to indicate their willingness for study participation using a pre-stamped envelope. A postal reminder was sent if no reply was received within four weeks. Patients who responded positively were contacted by telephone and met by a researcher at their home where written informed consent was obtained. This study was conducted together with another study investigating patients' understanding of information using patient interviews, the results of which are published elsewhere, 18 and the sample size calculation was based upon this study. Participants were asked to take part in both studies. Appropriate ethics committee approval was granted by the North Somerset and South Bristol Research Ethics Committee (project 07/H0106/185).

Patients' desire for information was assessed using the validated Toronto Information Needs Questionnaire, which was self-completed at patients' homes, without the presence of the researcher, and returned by post. Additional clinical and socio-demographic patient variables including age, gender, American Society of Anesthesiologists grade at the time of treatment, 19 tumour site and pre-treatment cancer stage 20 were recorded through case note review, and educational level and social class were obtained directly from patients. Educational level was defined as up to basic education (to the age of 16 or completion of the UK General Certificate of Secondary Education or equivalent), and further education (all subsequent education/qualifications). Social class was defined according to the UK Registrar-General's social class, 21 and classified into manual and non-manual.

Questionnaire properties

The Toronto Information Needs Questionnaire (TINQ) is a validated multi-dimensional 52 item instrument that measures patients' views on the importance of different types of information. 22 Item responses range from one (“not important”) to five (“extremely important”). The questionnaire has five scales assessing how important patients consider information about: 1) the disease (the nature, process and prognosis), 2) tests (procedures used to assess the extent of the disease, how and why they are done and sensations that may be experienced), 3) treatments (how they work and are performed, sensations that may be experienced and possible side effects), and quality of life outcomes including 4) physical effects of treatment and 5) psychosocial effects. The TINQ was originally designed to assess information needs of patients with breast cancer, therefore the four breast specific items were removed, producing a 48 item questionnaire. The questionnaire was scored according to the developers' instructions. 22 The five scale scores were converted to a 1–100 score for ease of interpretation.

Questionnaire selection was informed by a systematic review of cancer needs instruments. 23 The TINQ was chosen because it was the only detailed, multi-dimensional assessment tool of patients' information needs.

Statistical analysis

Analyses were undertaken to compare characteristics of questionnaire responders and non-responders using chi-square and t-tests.

Individual questionnaire item responses were initially analysed and categorized into the proportion of patients considering information as “very” or “extremely” important (hereafter defined as “high importance”) and compared to those who considered the information “not important”. Descriptive statistics were used to present individual items in order of importance to patients. Results are described in terms of the highest rated individual items overall, the highest rated items in each questionnaire scale, and the lowest rated individual items overall. Responses to other items are presented to illustrate selected concepts that were rated highly by patients but do not feature among the highest ranked items.

Mean values and 95 percent confidence intervals were also calculated for each of the five questionnaire scales. Scales with missing item responses were excluded from analyses. Scale distributions were examined to determine if parametric tests were appropriate. Univariable and multivariable linear regression was used to assess associations between scale scores and patient clinical and sociodemographic details. All measured patient characteristics described above were considered to be potential explanatory variables a priori, and were therefore included in the final multivariable model. Analyses were completed using STATA statistical software version 10.1 (StataCorp, College Station, TX). 24


Population and response rate

Two hundred and twenty six patients were invited to participate and 136 completed questionnaires, (response rate 60.2%, details: Table 1 ). Responders and non-responders were of similar age and had undergone similar types of treatment, however, responders were more likely to be women (p = 0.02), in the post-treatment phase (p = 0.001), and with junctionally located tumours (p = 0.03). Eight questionnaires were incomplete and led to an overall missing item rate of 28 out of 6528 (0.4%).

Table 1 Clinical and sociodemographic details of responders and non-responders. SD Standard deviation; ASA American Society of Anesthesiologists grade. 19

  Responders n = 136 Non-responders n = 90
Mean age (SD) 66 (8.8) 65 (13.0)
Male (%) 105 (77.2) 57 (63.6)
Tumour site (%)
 Oesophagus 77 (56.3) 57 (63.3)
 Oesophagogastric junction 26 (19.3) 6 (6.8)
 Stomach 33 (24.4) 27 (30.0)
Pre-treatment (%) 34 (25.0) 41 (45.6)
Treatment type (%)
 Surgery 34 (25.0) 21 (23.3)
 Chemotherapy and surgery 84 (61.8) 56 (62.2)
 Radical chemoradiotherapy 18 (13.4) 13 (14.4)
Tumour stage c (%)
 I 16 (11.8)  
 IIa 54 (39.7)  
 IIb 9 (6.62)  
 III 57 (41.9)  
ASA grade (%)
 1 38 (27.9)  
 2 68 (50.0)  
 3 30 (22.0)  
Educational level a (%)
 Basic 70 (51.5)  
 Further 66 (48.5)  
Social class b (%)
 Manual 63 (45.6)  
 Non-manual 73 (54.1)  

a Basic education is up to the age of 16 or completing the General Certificate of Secondary Education or equivalent. Further education is all other qualifications.

b UK Registrar-General’s Social Class (21), where I to IIIN is non-manual and IIIM to IV is manual.

c American Joint Committee on Cancer Staging system grouping(20).

Patients' information needs

A total of 35 (72.9 percent) individual questionnaire items were rated of high importance by more than 50% of patients (Table 2a and Table 2b). The highest overall rated individual items considered the chance of long-term survival, disease spread and recurrence and were rated highly important to 126 (92.7%), 125 (91.9%) and 120 (88.2%) patients respectively. All of these items were in the disease scale. Only 2 (1.5%), 1 (0.7%) and zero patients rated these items as not important.

Table 2a The proportion of patients rating single items as highly important and not important, together with mean item scores, in order of importance (continued overleaf).

Questionnaire item Scale Patients rating as high importance a Patients rating as not important
If there is cancer anywhere else in my body Disease 126 (92.7) 2 (1.5)
How to tell if the cancer has come back Disease 125 (91.9) 1 (0.7)
If the cancer will come back Disease 120 (88.2) 0
Who I should call if I have questions while I am still getting treatment Treatment 115 (84.6) 0
How the illness may affect my life over the next few months Disease 115 (84.6) 2 (1.5)
What side effects I should report to the doctor/nurse Treatment 113 (83.1) 1 (0.7)
If I am prone to infection because of my treatment Treatment 113 (83.1) 1 (0.7)
Why the doctor has suggested the treatment plan for me Treatment 113 (83.1) 0
How the illness could affect my life in the future Disease 112 (82.4) 2 (1.5)
If I have side effects, how to deal with them Treatment 112 (82.4) 2 (1.5)
How I will feel after my treatment Treatment 112 (82.4) 3 (2.2)
What treatment plans are available Treatment 110 (80.9) 2 (1.5)
If there are ways to prevent treatment side effects Treatment 109 (80.2) 2 (1.5)
The possible side effects of my treatment Treatment 106 (77.9) 3 (2.2)
If there will be changes in the usual things I can do with and for my family Psychosocial 104 (76.5) 3 (2.2)
Who I should call if I have questions after all the treatments are over Treatment 104 (76.5) 0
How the treatment works against the cancer Treatment 103 (75.7) 2 (1.5)
Which foods I can or cannot eat Physical 103 (75.7) 2 (1.5)
Where my family can go if they need help dealing with my illness Psychosocial 101 (74.3) 0
How to care for my wound or incision Physical 100 (73.5) 3 (2.2)
How long I will be receiving treatment Treatment 100 (73.5) 0
How to prepare for my treatment Treatment 100 (73.5) 5 (3.7)
The reasons why doctors suggest certain test Tests 94 (69.1) 4 (2.9)

a High importance was defined as rating “very” or “extremely” important to know on the five-point Likert scale. Mean scores are between one and five.

Table 2b The proportion of patients rating single items as highly important and not important, together with mean item scores, in order of importance (continued).

Questionnaire item Scale Patients rating as high importance a Patients rating as not important Mean score
How the cancer acts in the body Disease 94 (69.1) 3 (2.2) 3.9
If I'm going to need help taking care of myself Physical 92 (67.6) 3 (2.2) 3.9
What the results of the blood tests mean Tests 91 (66.9) 3 (2.2) 3.8
If there are physical things I should not do Physical 91 (66.9) 2 (1.5) 3.8
If it is known what causes intestinal cancer Disease 86 (63.2) 5 (3.7) 3.8
How my treatment is done Treatments 85 (62.5) 3 (2.2) 3.8
If my illness is hereditary Disease 79 (58.1) 12 (8.8) 3.6
What to do if I become concerned about dying Psychosocial 75 (55.1) 10 (7.4) 3.6
If I can take a bath or shower Physical 75 (55.1) 9 (6.6) 3.5
How to prepare for the tests Tests 75 (55.1) 6 (4.4) 3.6
Who to talk with if I hear about treatments apart from surgery, radiation, or chemotherapy Treatments 69 (50.7) 6 (4.4) 3.5
How long will my incision/wound take to heal Physical 69 (50.7) 6 (4.4) 3.5
Why they need to test my blood Tests 64 (47.1) 6 (4.4) 3.4
How I will feel after the tests Tests 64 (47.1) 6 (4.4) 3.4
Can I continue my usual social activities Physical 62 (45.6) 8 (5.9) 3.3
How to talk to my family/friends about my illness Psychosocial 61 (44.9) 10 (7.4) 3.3
Where I can get help to deal with my feelings about my illness Psychosocial 60 (44.1) 8 (5.9) 3.3
How I will feel during the tests Tests 59 (43.4) 15 (11.0) 3.3
If I can continue my usual hobbies and sports Physical 58 (42.6) 6 (4.4) 3.4
How the tests are done Tests 52 (38.2) 9 (6.6) 3.2
What to do if I feel uncomfortable in social situations Psychosocial 48 (35.3) 22 (16.2) 2.9
If there are any groups where I can talk with other people who have cancer Psychosocial 44 (32.4) 20 (14.7) 2.9
The medical name for my type of cancer Disease 43 (31.6) 16 (11.8) 3.0
If the treatment will alter the way I look Treatment 38 (27.9) 26 (19.1) 2.8
Where I can get help if I have problems feeling as attractive as I did before Psychosocial 32 (23.5) 36 (26.5) 2.5

a High importance was defined as rating “very” or “extremely” important to know on the five-point Likert scale. Mean scores are between one and five.

The highest rated individual items in the other scales were “Who I should call if I have questions while I am still getting treatment” (treatment scale), “If there will be changes in the usual things I can do with and for my family” (psychosocial scale), “Which foods I can or cannot eat” (physical scale) and “The reasons why doctors suggest certain tests” (tests scale), which were considered highly important by 115 (84.6%), 104 (76.5%) 103 (75.7%), and 94 (69.1%) patients respectively.

The lowest rated information items were “Where I can get help if I have problems feeling as attractive as I did before” (psychosocial scale), “If the treatment will alter the way I look” (treatment scale) and “The medical name for my type of cancer” (disease scale), although all these were still considered highly important by more than 32 (23.5%) patients and rated not important by fewer than 36 (26.5%).

Other concepts that were rated highly included the need to understand treatment rationale (e.g. “why the doctor has suggested the treatment plan for me”, with 113 (83.1%) patients rating this as highly important) and quality of life concepts (“how the illness may affect my life in the future” and “how I will feel after my treatment”, both considered highly important by 112 (82.4%) patients).

Predictors of information needs

The highest mean scale scores were for information about the disease (76.3, 95% CI 73.6–79.1) and treatments (74.8, 95% CI 72.1–77.5, Table 3 ). Information about tests, physical and psychosocial effects obtained mean scores of 63.3, 67.2 and 58.2 respectively.

Table 3 Mean and median scale scores, 95% confidence intervals, standard deviations and skewness. CI confidence interval; SD standard deviation.

Scale Mean 95% CI Median SD Skewness
Disease 76.3 73.6–79.1 77. 8 16.3 −0.93
Treatments 74.8 72.1–77.5 73.4 15.6 −0.36
Physical effects 67.2 64.1–70.2 68.8 17.9 −0.82
Tests 63.3 59.5–67.0 60.7 21.8 −0.56
Psychosocial effects 58.2 54.7–61.6 56.3 19.8 −0.25

Measures of central tendency demonstrated that scale scores were normally distributed ( Table 3 ). Univariable linear regression demonstrated evidence to suggest that women rated information relating to all five scales as more important than men; in particular information about treatments, tests and psychosocial effects with mean differences in scores of 9.1 (95% CI 2.9–15.4, p = 0.005), 11.1 (95% CI 2.3–19.9, p = 0.013) and 12.1 (95% CI 4.1–20.0, p = 0.003) respectively. There was also weak evidence of certain patterns across the five scales. In general, non-manual social class, further education, ASA grade 2/3, and tumour stage IIa-III led to information relating to all scales being deemed less important, however, the P values demonstrate that these patterns are likely to be explained by chance ( Table 4 ). In terms of tumour site, compared to oesophagus, patients with oesophago-gastric junction cancer reported less importance for all five scales and patients with gastric more, but this again may be explained by chance.

Table 4 Univariable linear regression assessing associations between scale scores and clinical and sociodemographic patient details.

  Disease Treatments Physical Tests Psychosocial
Coef. 95% CI P value Coef. 95% CI P value Coef. 95% CI P value Coef. 95% CI P value Coef. 95% CI P value
Age (per unit) −0.1 (−0.4 to 0.2) 0.5 −0.1 (−0.4 to 0.2) 0.5 0.1 (−0.3 to 0.5) 0.6 −0.1 (−0.6 to 0.3) 0.6 −0.3 (−0.7 to 0.1) 0.2
Social class c
Non-manual a 0     0     0     0     0    
Manual 3.3 (−2.3 to 8.9) 0.2 0.9 (−4.6 to 6.3) 0.8 1.3 (−4.8 to 7.5) 0.7 4.1 (−3.3 to 11.6) 0.3 5.9 (−1.1 to 12.9) 0.1
Educational level b
Basic a 0     0     0     0     0    
Further −1.7 (−7.3 to 3.9) 0.6 0.1 (−5.3 to 5.5) 0.9 −1.5 (−7.7 to 4.7) 0.6 −1.9 (−9.4 to 5.5) 0.6 −2.0 (−9.0 to 5.1) 0.6
Grade 1 a 0   0.1 0   0.1 0   0.07 0   0.4 0   0.3
Grade 2 −5.9 (−12.4 to 0.7)   −6.4 (−12.7 to 0.0)   −8.2 (−15.3 to −1.0)   −5.6 (−14.4 to 3.2)   −6.2 (−14.2 to 2.0)  
Grade 3 −0.6 (−8.5 to 7.2)   −2.1 (−9.7 to 5.4)   −3.1 (−11.8 to 5.6)   −1.1 (−11.7 to 9.5)   −4.8 (−14.7 to 5.2)  
Tumour stage d
I a 0   0.3 0   0.2 0   0.2 0   0.7 0   0.99
IIa −8.6 (−17.8 to 0.5)   −8.6 (−17.6 to 0.5)   −4.4 (−14.5 to 5.6)   −1.6 (−14.0 to 10.8)   −1.3 (−13.0 to 10.5)  
IIb −2.2 (−15.6 to 11.2)   −4 (−17.1 to 9.0)   −10.6 (−25.3 to 4.1)   1.6 (−16.5 to 19.7)   −2.6 (−19.4 to 14.2)  
III −5.3 (−14.4 to 3.8)   −8.2 (−17.2 to 0.8)   −9.0 (−19.0 to 1.0)   −4.9 (−17.3 to 7.4)   −0.9 (−12.6 to 10.8)  
Tumour site
Oesophagus a 0   0.08 0   0.2 0   0.1 0   0.4 0   0.4
OGJ −2.5 (−9.8 to 4.7)   −2.9 (−10.0 to 4.2)   −5.1 (−13.2 to 3.0)   −3.0 (−12.8 to 6.8)   −5.1 (−14.6 to 4.3)  
Stomach 6.3 (−0.4 to 12.9)   4.4 (−2.1 to 10.9)   4.2 (−3.1 to 11.6)   4.9 (−4.1 to 13.9)   2.7 (−5.8 to 11.2)  
Female sex 7.9 (1.3 to 14.6) 0.02 9.1 (2.9 to 15.4) 0.005 6.2 (−1.1 to 13.5) 0.095 11.1 (2.3 to 19.9) 0.013 12.1 (4.1 to 20.0) 0.003
Treatment phase
Post-treatment a 0   0.7 0   0.9 0   0.3 0   0.5 0   1.0
Pre/during treatment −1.3 (−7.7 to 5.1)   −0.5 (−6.6 to 5.7)   4.0 (−3.1 to 11.0)   −3.0 (−11.5 to 5.5)   0.2 (−7.8 to 8.2)  
Treatment type
Surgery a 0   0.6 0   0.2 0   0.7 0   0.2 0   0.3
Surgery and chemotherapy −2.2 (−8.8 to 4.4)   −2.5 (−8.8 to 3.9)   −3 (−10.2 to 4.3)   1.2 (−7.5 to 9.9)   −0.2 (−8.3 to 8.6)  
Radical chemoradiotherapy 1.5 (−8.4 to 11.3)   4.9 (−4.3 to 14.2)   −3.7 (−14.5 to 7.1)   10.9 (−1.9 to 23.7)   8.1 (−4.1 to 20.2)  

a Category denoted as the reference group to which all others compared.

b Basic education is up to the age of 16 or completing the General Certificate of Secondary Education or equivalent. Further education is all other qualifications.

c UK Registrar-General’s Social Class(21), where I to IIIN is non-manual and IIIM to IV is manual.

d American Joint Committee on Cancer Staging system grouping(20).

Coef Coefficient; CI Confidence interval; ASA American Society of Anaesthesiology; OGJ oesophago-gastric junction.

In the multivariable model the effect of gender remained, although attenuated. The difference in mean scale scores between women and men was 7.6 (95% CI 0.4–14.7, p = 0.038) for treatment, 8.8 (95% CI −1.2 to 18.8, p = 0.09) for test and 10.6 (95% CI 1.3–19.9, p = 0.025) for psychosocial effect information ( Table 5 ). The effect of social class was slightly augmented for all five scales. Patterns observed for ASA grade and tumour site remained similar to those observed in the univariable models, but the strength of evidence remained weak. The effect of education level disappeared in the multivariable model.

Table 5 Multivariable linear regression assessing associations between scale scores and clinical and sociodemographic patient details.

  Disease Treatments Physical Tests Psychosocial
Coef. 95% CI P value Coef. 95% CI P value Coef. 95% CI P value Coef. 95% CI P value Coef. 95% CI P value
Age (per unit) 0 (−0.4 to 0.4) 0.9 0 (−0.3 to 0.4) 0.8 0.3 (−0.1 to 0.8) 0.1 0 (−0.6 to 0.5) 0.9 −0.2 (−0.6 to 0.3) 0.5
Social class b
Non-manual a 0     0     0     0     0    
Manual 5.7 (−0.0 to 11.5) 0.05 3.9 (−1.6 to 9.4) 0.2 3.2 (−3.1 to 9.6) 0.3 7.4 (−0.3 to 15.1) 0.1 9.1 (1.9 to 16.4) 0.014
Educational level d
GCSE or below a 0     0     0     0     0    
Above GCSE 0.8 (−5.2 to 6.9) 0.8 2.2 (−3.5 to 8.0) 0.4 0.5 (−6.1 to 7.2) 0.9 0.2 (−7.8 to 8.3) 1 −0.7 (−8.2 to 6.8) 0.9
Grade 1 a 0   0.2 0   0.1 0   0.053 0   0.3 0   0.4
Grade 2 −6.1 (−13.4 to 1.1)   −7.0 (−13.9 to 0.0)   −9.8 (−17.8 to -1.9)   −7.7 (−17.4 to 2.0)   −6.3 (−15.3 to 2.7)  
Grade 3 −2.7 (−12.0 to 6.6)   −3.9 (−12.7 to 5.0)   −6.6 (−16.9 to 3.7)   −5.7 (−18.1 to 6.7)   −6.3 (−17.8 to 5.2)  
Tumour stage c
I a 0   0.5 0   0.4 0   0.4 0   0.6 0   0.96
IIa −6 (−15.8 to 3.8)   −6.9 (−16.5 to 2.6)   −3.2 (−14.0 to 7.7)   0.9 (−12.2 to 14.1)   1.9 (−10.3 to 14.4)  
IIb −0.6 (−14.5 to 13.3)   −3.3 (−16.7 to 10.2)   −6.9 (−22.2 to 8.4)   1.5 (−17.1 to 20.1)   0.4 (−16.7 to 17.4)  
III −2.8 (−12.9 to 7.3)   −7.7 (−17.4 to 2.1)   −8.2 (−19.3 to 2.9)   −5.1 (−18.6 to 8.5)   0.4 (−12.2 to 12.9)  
Tumour site
Oesophagus a 0   0.2 0   0.4 0   0.3 0   0.1 0   0.2
OGJ −0.6 (−8.2 to 7.1)   −1 (−8.3 to 6.4)   −5.6 (−14.1 to 2.8)   1.7 (−8.5 to 11.9)   −1.9 (−11.6 to 7.8)  
Stomach 7.8 (−1.0 to 16.5)   5 (−3.4 to 13.4)   1.3 (−8.3 to 10.9)   11.6 (−0.1 to 23.3)   8.5 (−3.0 to 20.0)  
Female sex 7.1 (−0.5 to 14.6) 0.066 7.6 (0.4 to 14.7) 0.038 6.8 (−1.5 to 15.1) 0.11 8.8 (−1.2 to 18.8) 0.09 10.6 (1.3 to 19.9) 0.025
Treatment phase
Post-treatment a 0     0     0     0     0    
Pre/during treatment −1.8 (−8.9 to 5.2) 0.6 −0.7 (−7.5 to 6.1) 0.8 5.5 (−2.3 to 13.2) 0.2 −5 (−14.4 to 4.4) 0.3 −1.7 (−10.4 to 7.1) 0.7
Treatment type
Surgery a 0   0.8 0   0.35 0   0.96 0   0.076 0   0.4
Surgery and chemotherapy 1.1 (−7.5 to 9.7)   0.7 (−7.5 to 9.0)   −0.7 (−10.2 to 8.7)   8.5 (−3.0 to 20.0)   1.7 (−9.5 to 12.9)  
Radical chemotherapy 4.1 (−8.0 to 16.1)   7.2 (−4.3 to 18.7)   −1.8 (−15.3 to 11.7)   18.6 (2.5 to 34.6)   9.2 (−6.3 to 24.7)  

a Category denoted as the reference group to which all others compared.

b UK Registrar-General’s Social Class(21), where I to IIIN is non-manual and IIIM to IV is manual.

c American Joint Committee on Cancer Staging system grouping(20).

d Basic education is up to the age of 16 or completing the General Certificate of Secondary Education or equivalent. Further education is all other qualifications.

Coef Coefficient; CI Confidence interval; ASA American society of anaesthesiology; OGJ oesophago-gastric junction.


This survey of information needs of patients undergoing curative OG cancer treatment showed that most information was considered “highly important”, especially information about prognosis which was considered of high importance by over 90% of patients. Other concepts that patients rated highly, included information about the rationale behind treatment recommendations, and information about quality of life after treatment. The least important information item related to physical appearance, but this was still of high importance to over a quarter of patients. Multivariable regression did not identify any strong evidence of associations between desire for information and patient clinical and socio-demographic characteristics, although data suggested that women rate information relating to treatment and psychosocial effects as more important than men, but due to the large number of statistical tests performed these may be chance findings.

Results from this study are in line with research assessing information needs in other populations of patients with different cancers. Some of the evidence is summarized in a systematic review that demonstrated cancer patients' widespread desire for treatment-related information (for example, risks and benefits of treatment and treatment options), rehabilitation (including self-care issues and long-term side effects) and prognosis, although this does not focus specifically on populations of patients with OG cancer. 15 Four subsequent studies have investigated this issue in patients with upper gastro intestinal cancer, however, these studies either had a poor sample size or were limited to only investigating desire for prognostic information.25, 26, 27, and 28 In a study of over 170 patients with oesophageal cancer, 97 percent wanted realistic prognostic information and, importantly, the majority wanted doctors to initiate discussions about long-term survival. 26 Another study which used discrete-choice experiment methods examined OG cancer patients' preferences for different treatment outcomes. 29 Results showed that although cure was important, it was surpassed by a desire for good quality of life. Combined with these studies of patients with other cancers, our results therefore confirm the importance of clinicians sensitively and skillfully addressing the issues of survival and prognosis when discussing treatment for oesophageal and gastric cancer. They should be aware that patients may be reluctant to ask even if they want the information because they fear what they may be told.

Identifying and surveying patients undergoing OG cancer treatment is challenging because of the severity of the disease and poor treatment outcomes. This may account for the observed 40 percent non-response rate in this study, leading to potential selection bias. Indeed, selection bias may have resulted in overestimation of patients' desire for information, because those with a low desire for information may have been disinclined to participate. Alternatively, potential selection bias may have led to under estimation, because unwell patients with a high desire for information may not have participated because the survey process was onerous. Non-responders were demonstrated to have similar characteristics to responders based on a limited number of variables and is provides some evidence against systemic bias. Data collection on non-responders was limited by research ethics regulation and all available data is presented in this report. Although it is not clear how selection bias may have influenced the study findings, these data represent some of the only evidence of information needs of patients with OG cancer. The majority of patients that participated in this study had already completed treatment and, although this provides the opportunity to reflect on their desires with hindsight, results may not be fully relevant to patients that are about to undergo curative treatment. Similarly, by grouping results of pre- and post-treatment patients together, preferences for the timing of information are not identified.

Patients' reported information needs may have been influenced by societal pressure, however, several mechanisms were used to minimize these effects. Firstly, patients were informed that their answers would be fully anonymised and they were free to answer questions based on personal feelings. Secondly, patients were asked to complete questionnaires in their private home at a time of their choosing, so to reduce any perceived pressure from external bodies such as health services. Finally, the questionnaires were distributed from the university and not from the clinical team involved with the patients care, so to potentially reduce any desire to “please” treating clinicians.

Information needs of patients with OG cancer are important because outcomes are poor and treatments highly morbid. This study surveyed patients with oesophageal and gastric cancer together because surgery results in mainly similar postoperative risks, outcomes and health-related quality of life profiles. There are, however, some important distinctions between these groups of patients that may have influenced the study findings. Differences in disease aetiology may have affected the participants' personal characteristics. For example, oesophageal squamous cell cancer is associated with lower socioeconomic status compared to adenocarcinoma of the oesophagus and stomach. This was, however, considered by controlling for this potential confounder a priori in the analysis.

This paper provides evidence that most patients with OG want detailed information including prognostic chances and quality of life outcomes. It is therefore recommended that clinicians consider communicating these outcomes during patient consultations to improve patient-centered care. It would be incorrect, however, to simply provide all patients with uniform information. All individuals do not want all information and it would be inappropriate and potentially unethical to impose unwanted information upon them. 30 In addition, even if patients want detailed information, communicating large amounts routinely at one place and time may result in information “overload” and may reduce patient understanding. One solution to this problem would be to ask each individual what information they would like to know, but many may not have the initial understanding to articulate what information is desired. We therefore propose that “core disclosure” is the optimal solution to providing patients with the correct amount of information. Core disclosure represents a minimum amount of information that is initially communicated to stimulate patients to consider what else they would like to know. 30 It therefore acts as a baseline that both prevents information overload and allows individually tailored information provision. Future research is required to define what information patients with OG cancer consider is “core”, as well as to test impact of core disclosure on consultations in a large multi-centre setting and the effects of information on patient sub-groups. This study does, however, provide some evidence for the types of information that would be included in a core disclosure set.


The authors would like to thank the oesophago-gastric multi-disciplinary team at University Hospitals Bristol NHS Foundation Trust for their help and support, and the Royal College of Surgeons of England for financial support. The study sponsors had no involvement in the study design, in the collection, analysis and interpretation of data; in the writing of the manuscript; and in the decision to submit the manuscript for publication.

Conflict of interest


Source of support

AGKM was supported by the Royal College of Surgeons Sir Alan Parks Research Fellowship.


  • 1 L.D. Baum. Psychosocial needs of patients with cancer in the primary care setting. Lippincotts Prim Care Pract. 2000;4(4):417-425
  • 2 J. Burt, K. Caelli, K. Moore, M. Anderson. Radical prostatectomy: men's experiences and postoperative needs. J Clin Nurs. 2005;14(7):883-890
  • 3 P.N. Butow, R.F. Brown, S. Cogar, M.H. Tattersall, S.M. Dunn. Oncologists' reactions to cancer patients' verbal cues. Psychooncology. 2002;11(1):47-58
  • 4 R.M. Epstein, R.L. Street. Patient-centered communication in cancer care: promoting healing and reducing suffering. (National Cancer Institute, Bethesda, MD, 2007) NIH Publication No. 07-6225
  • 5 C. Lerman, M. Daly, W.P. Walsh, et al. Communication between patients with breast cancer and health care providers. Determinants and implications. Cancer. 1993;72(9):2612-2620
  • 6 G.M. Leydon, M. Boulton, C. Moynihan, et al. Cancer patients' information needs and information seeking behaviour: in depth interview study. BMJ. 2000;320:909-913
  • 7 B. van der Molen. Relating information needs to the cancer experience: 1. information as a key coping strategy. Eur J Cancer Care (Engl). 1999;8(4):238-244
  • 8 S. Chaitchik, S. Kreitler, S. Shaked, I. Schwartz, R. Rosin. Doctor-patient communication in a cancer ward. J Cancer Educ. 1992;7(1):41-54
  • 9 L. Fallowfield, V. Jenkins. Effective communication skills are the key to good cancer care. Eur J Cancer. 1999;35(11):1592-1597
  • 10 P.J. Moore, N.E. Adler, P.A. Robertson. Medical malpractice: the effect of doctor-patient relations on medical patient perceptions and malpractice intentions. West J Med. 2000;173(4):244-250
  • 11 L.M. Ong, M.R. Visser, F.B. Lammes, J.C. de Haes. Doctor-patient communication and cancer patients' quality of life and satisfaction. Patient Educ Couns. 2000;41(2):145-156
  • 12 R.L. Street Jr., G. Makoul, N.K. Arora, R.M. Epstein. How does communication heal? Pathways linking clinician-patient communication to health outcomes. Patient Educ Couns. 2009;74(3):295-301
  • 13 Department of Health. Better information, better choices, better health: putting information at the centre of health. (Department of Health, London, 16-12-2004) [Ref Type: Report]
  • 14 The royal college of surgeons of England. good surgical practice (RCSENG – Professional Standards and Regulation, 2008)
  • 15 L.J. Rutten, N.K. Arora, A.D. Bakos, N. Aziz, J. Rowland. Information needs and sources of information among cancer patients: a systematic review of research (1980–2003). Patient Educ Couns. 2005;57(3):250-261
  • 16 K. Avery, R. Hughes, A. McNair, D. Alderson, P. Barham, J. Blazeby. Health-related quality of life and survival in the 2 years after surgery for gastric cancer. Eur J Surg Oncol. 2010;36(2):148-154
  • 17 P.C. Enzinger, R.J. Mayer. Esophageal cancer. N Engl J Med. 2003;349(23):2241-2252
  • 18 A.G. McNair, S.T. Brookes, C.R. Davis, M. Argyropoulos, J.M. Blazeby. Communicating the results of randomized clinical trials: do patients understand multidimensional patient-reported outcomes?. J Clin Oncol. 2010;28(5):738-743
  • 19 C.J. Vacanti, R.J. van Houten, R.C. Hill. A statistical analysis of the relationship of physical status to post operative mortality in 68,388 cases. Anaesth Analgesia. 1970;49:564
  • 20 F.L. Greene, D.L. Page, I.D. Fleming, et al. AJCC cancer staging manual. (Springer-Verlag, New York, NY, 2002)
  • 21 OPCS. Classification of occupations. (HMSO, London, 1980)
  • 22 S. Galloway, J. Graydon, D. Harrison, et al. Toronto information needs questionnaire-breast cancer. B.K. Redman (Ed.) Measurement tools in patient education (Springer Publishing, New York, 2003) 249-255
  • 23 K.Y. Wen, D.H. Gustafson. Needs assessment for cancer patients and their families. Health Qual Life Outcome.. 2004;2:11
  • 24 StataCorp. Stata/IC. [10.1]. 2008. (, 1985) Texas, USA
  • 25 S. Andreassen, I. Randers, E. Naslund, D. Stockeld, A.C. Mattiasson. Information needs following a diagnosis of oesophageal cancer; self-perceived information needs of patients and family members compared with the perceptions of healthcare professionals: a pilot study. Eur J Cancer Care. 2007;
  • 26 S.J. Franssen, S.M. Lagarde, J.R. van Werven, et al. Psychological factors and preferences for communicating prognosis in esophageal cancer patients. Psychooncology. 2009;18(11):1199-1207
  • 27 S.M. Lagarde, S.J. Franssen, W. van Jr., et al. Patient preferences for the disclosure of prognosis after esophagectomy for cancer with curative intent. Ann Surg Oncol. 2008;
  • 28 E. Wittmann, C. Beaton, W.G. Lewis, et al. Comparison of patients' needs and doctors' perceptions of information requirements related to a diagnosis of oesophageal or gastric cancer. Eur J Cancer Care (Engl). 2010;
  • 29 S.G. Thrumurthy, J.J. Morris, M.M. Mughal, J.B. Ward. Discrete-choice preference comparison between patients and doctors for the surgical management of oesophagogastric cancer. Br J Surg. 2011;98(8):1124-1131
  • 30 T.L. Beauchamp, J.F. Childress. Principles of biomedical ethics. (Oxford University Press, Oxford, 2001)


a Centre for Surgical Research, School of Social and Community Medicine, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol BS8 2PS, UK

b Division of Surgery, Head and Neck, University Hospitals Bristol NHS Foundation Trust, Bristol BS2 8HW, UK

c Institute of Primary Care and Public Health, School of Medicine, Cardiff University, Heath Park, Cardiff CF14 4XN, UK

Corresponding author. School of Social & Community Medicine, Canynge Hall, 39 Whatley Road, Clifton, Bristol BS8 2PS, UK. Tel.: +44 (0) 117 928 7231; fax: +44 (0) 117 928 7292.

Search this site

Stay up-to-date with our monthly e-alert

If you want to regularly receive information on what is happening in Quality of Life in Oncology research sign up to our e-alert.

Subscribe »

QOL (Quality of Life) newsletter e-alert

NEW! Free access to the digital version of a new publication in Cancer Supportive Care

Cancer cachexia: mechanisms and progress in treatment

Authors: Egidio Del Fabbro, Kenneth Fearon, Florian Strasser

This book was supported by an educational grant from Helsinn Healthcare SA.

Featured videos

Quality of Life promotional video

Made possible by an educational grant from Helsinn

Helsinn does not have any influence on the content and all items are subject to independent peer and editorial review

Society Partners

European Cancer Organisation Logo