Supporting Doctor-patient Communication in Oncology

Methods/study plan

1. Translation of QPL to Norwegian
2. Reference group
3. Pilot project
4. RCT
5. Translation of Question Prompt List (QPL) and evaluation of Norwegian version This QPL consists of a list of 49 questions covering 9 topics and is not yet used in Norway nor translated into Norwegian. Our collaborating partners , professor Phyllis Butow and professor Martin Tattersal at the Centre for Medical Psychology and Evidence-based Decision-making at the University of Sydney have developed several QPLs for different settings. In collaboration with prof. Butow and colleagues we have selected the QPL "Questions to ask your medical or radiation oncologist" which can be considered as suitable for the consultation at the outpatient oncology clinic at the University hospital of North Norway and will be translated to Norwegian.

The translation of the QPL will be done in 5 steps using the European organization for research and treatment of cancer (EORTC) gold-standard approach

1. Two independent translators with Norwegian as their mother tongue translate QPL from English to Norwegian.
2. Merge of the two translations after discussion between the two translators and the study coordinator.
3. Back translation into English of an independent third translator.
4. Discussion of the translation in a committee of researchers and clinicians. The versions agreed on will be the prefinal version. The Norwegian version of QPL The Norwegian QPL must be culturally sensitive for Norwegian speaking cancer patients and there may be questions in the Australian QPL that are not relevant for Norwegian conditions. The investigators will arrange approximately 3 focus groups of 6-10 experienced cancer patients or former cancer patients to provide information of any questions that are lacking in the Australian version and questions that might be excluded in the Norwegian version. Patients who participated in the nationwide survey "Outpatient experiences with cancer clinics in Norway 2011" reported lack of information about side effects of treatment, and lack of information about pain / pain relief. Due to those results the researchers want to discuss weather to add some questions to the QPL, regarding side effects of treatment and pain/ pain treatment with members of the focus group.
5. Discussion of the information retrieved from the focus groups in a committee of researchers and clinicians and changes might be done on the prefinal version.
6. Test the prefinal Norwegian version (that might have been altered after the focus groups) on approximately 30 experienced cancer patients. The patients are invited to give their opinion on each of the prompted questions. The investigators will ask them to grade all the questions according to:
7. Usefulness 2. Clarity 3. The amount of discomfort such a question would generate The patients will also be asked to suggest additional questions if important issues are missing.

The experienced cancer patients in 4) and 6) will be recruited from Vardesenteret.

In addition, the investigators want to make a small change in the design by leaving space in front of the prompted questions for patients to make their own questions as suggested in a study by Voltz.

1. Reference group There may be culturally differences between Australian and Norwegian cancer patients. As studies with use of QPLs and audio recording have not been done with Norwegian cancer patients, the investigators need a reference group to study how Norwegian cancer patients in general participate during the first consultation with the oncologist and to what extend Norwegian patients are involved in decision making. The investigators plan to record 30 consecutive first time consultations with cancer patients not receiving QPL and have this as a reference group and count questions and issues that are discussed in the consultation and code the consultation according to Option scale. They will also be asked to fill inn one questionnaire of baseline characteristics and information retrieved. The reference group will be necessary to be able to calculate sample size of participants needed in the RCT. The patients will receive written information before their consultation and will also be contacted by phone prior to the consultation to agree to participate and sign a written consent.
2. Pilot study (Interim analyses) Before starting an RCT the study concept will be tested in a pilot study. These patients will receive written information before their consultation and will also be contacted by phone prior to the consultation. 30 consecutive cancer patients at their first consultation at the outpatient clinic will be provided with a QPL and an audio file of the consultation. These patients will be encouraged from their oncologist to ask questions after a standard statement. They will also receive the same questionnaires which will be used in the RCT (see below).

The investigators will also conduct training of the oncologists prior to the pilot study to ensure that they encourage the use of QPL in a standard manner.

1. RCT This is a single center, randomized control trial at the oncology outpatient clinic at the University hospital North Norway. The clinic is situated as a separate part of the cancer department and is served by specialists in oncology and junior doctors in training. The consultation length is usually about 40-60 minutes for first time consultations.

Participants - Study design All patients who are referred to the oncology outpatient clinic and meet the inclusion criteria are invited to participate in the study along with the letter for their first appointment. There are approximately 15 new referrals every week.

If agreeing to participate and having signed a written informed consent, patients will be randomized into intervention or control arm (see below). Participants in both groups are asked to fill out questionnaires to detect baseline characteristics, anxiety/depression and their wish for shared decision-making (control preference scale).

Patients in the intervention group will receive the QPL at home in advance of the consultation along with information about the clinic.

All specialists in oncology and also junior physicians with at least one years of training in oncology are invited to participate. The doctors will be instructed to address the QLP in a standardized way as early in the consultation as possible. They will also be asked to encourage use of QPL on further consultation within the two month period, as the patients will be asked to fill out questionnaires (see below) one week and two months after the first consultation.

The consultations will be recorded in both the control and intervention group. The recording will be done on the computer, and the patients in the intervention group will be given the recording immediately after the consultation on a memory stick. For both groups the audiotapes will be analyzed after standard procedures (see below).

Endpoint in RCT Primary endpoint Difference in number of questions asked, and especially concerning prognosis. Secondary endpoint Difference in shared decision making. Difference in anxiety/depression/health related quality of life. Difference in satisfaction with the consultation and information retrieved.

Analysis of the audio recorded consultations Our collaborators in Sydney have developed a manual and coding sheet to enable standardization of the coding procedure for the consultations with "Question prompt lists".This will be translated to Norwegian. The audio recordings will be transcribed and patients questions and concerns will be counted and categorized according to topics covered by the QPL. Physician endorsement of the QPL will be coded as either absent, basic or extended.

The coders will also be trained to use the "Option scale" to score patient involvement in decision making. The option scale consists of 12 dimensions of a collaborative decision making process and the dimensions will be registered by frequency.

The coders will be blinded to group allocation. Intra-rater and inter-rater reliability will be tested. Training of coders will be under the guidance of Prof. Svein Bergsvik in collaboration with Prof. Butow. The coders will probably be 2nd year students in psychology who can follow the project for the three year period and will be paid on an hourly basis.

Sample size calculations To calculate sample size for the study, the investigators need to know average numbers of questions asked by Norwegian cancer patients in general. The investigators will get this information from the reference group. In the absence of data from Norway, the investigators assume it will be the same as from Australia were studies report a mean number of 9 questions (range 0-53). A sample size calculation suggest that we need n=80 patients in each group in order to have 80% power to detect a significance difference between the groups when we anticipate a mean difference between the groups of 3 and a standard deviation of 5 using a 5% significance level.

Randomization Patients will be randomly assigned and stratified for age, sex, education, type of cancer (breast cancer vs non-breast cancer) and curative or palliative treatment intention. The randomization will be done with a concealed, computer- generated randomization procedure. As the same doctor will meet patients in both the intervention and the control arm, this must be taken into consideration either in the randomization or the statistical analyzes or both.

Statistical analyses Analysis will be by intention to treat, using all available data from randomly assigned patients according to group assignment. Comparison between groups will be performed using chi-square tests or logistic regression for binary variables and with two sample t-test or linear regression models for continuous variables. Data will be analyzed using the SPSS version 19 (SPSS Inc, Chicago, Illinois) and SAS version 9.2 (SAS Institute, Cary, NC).