Author's response to reviews

Transcription

1 Author's response to reviews Title: What determines patient preferences for treating basal cell carcinoma? A discrete choice experiment survey from the SINS trial (surgery vs imiquimod) Authors: Michela Tinelli (M.Tinelli@lse.ac.uk) Mara Ozolins (Mara.Ozolins@Nottingham.ac.uk) Fiona J Bath-Hextall (Fiona.Bath-Hextall@Nottingham.ac.uk) Hywel C Williams (Hywel.Williams@Nottingham.ac.uk) Version: 2 Date: 30 August 2012 Author's response to reviews: see over

2 MS: Research article What determines patient preferences for treating basal cell carcinoma? A discrete choice experiment survey from the SINS trial (surgery vs imiquimod) Michela Tinelli, Mara Ozolins, Fiona J Bath-Hextall and Hywel C Williams BMC Dermatology Thank you for your reviewers comments which have helped us to improve our manuscript. Please find our responses in the table below, including reference to changes in the manuscript. We hope you will be satisfied with the revised manuscript, but please get back to us if you require any further changes, and we will be happy to respond. Reviewer - Brigitte Essers 1 The authors state that they used attributes and levels from a previous DCE applied to BCC and from discussions with the research team and experts. However, I think that this is too concise. Have the authors conducted focus groups (with patients) to gain insight into those treatment characteristics that are considered important by patients? Were the discussions with the research team and experts conducted in a systematic manner? Many thanks for your comment. Please see amended text as follow: Page 6- methods In this experiment, a list of attributes, their levels, and status quo alternative (see below) were derived from a previous DCE exercise applied to BCC (Weston et al [2]), and from systematic discussion with the research team plus advice provided by a panel of experts. Page 16, discussion: Time and budget constraints did not allow us to further inform the decision on the choice of the attributes and their levels with patient interviews and focus groups. 2 Why was the administration and time period of each treatment not included as an attribute? A difference between surgical excision and imiquimod is that surgical excision is performed on one and the same day in a hospital setting while imiquimod has to be applied by the patient over a number of weeks in his/her home environment. I can imagine that this could be an important attribute for patients when indicating their preference. Unfortunately limited wording did not allow us to include further details on the questionnaire and how the treatment/attributes were presented. The surgery treatment was presented as a once only treatment that requires attendance at the hospital outpatients department, whilst the cream treatment was presented as a treatment that is applied by you, at home, once a day for a period of 6 to 12 weeks. This allowed us to keep the number of attributes, and therefore choices, to a minimum. We also stated that if the cream fails to work you will require surgery and if the surgery fails to work you will require further surgery. See amended text p6 For further

3 3 How was the cost-attribute explained to patients? I assume that in reality treatment for skin cancer is free of charge. 4 The levels of the cost-attribute for surgical excision and imiquimod are similar. Is this accurate? One would expect surgical excision to be more expensive. Hence, the range for the cost-attribute levels would be different between surgical excision and imiquimod. 5 How was the attribute chance of complete clearance explained? Was the time-period also defined and explained to patients? Are the percentages clearance based on threemonths or twelve months follow-up? I guess the same is true for appearance. Is this the appearance three months after treatment, six months or one year after treatment? 6 Both patients with nodular or superficial BCC were included. Is it possible to perform a subgroup analysis on type of BCC (nodular and superficial)? details on how the treatments and attributes were presented to the patients please see the questionnaire in appendix. The cost to you attribute was added to find out how much patients valued the treatment (Please see amended text, page 6) although we assured them that the treatments compared were still due to be provided free of charge on the NHS (see questionnaire). The cost attribute levels are hypothetical and aimed to inform the DCE benefit evaluation exercise, alongside the formal trial economic evaluation. See amended text, page 16 Other attribute levels (e.g. different levels for the cost attributes depending on the specific treatment compared) could be considered for future exercises and informed by the trial economic evaluation. The chance of complete clearance was presented as the probability that BCC is completely removed (followed by an actual example). Both clearance and appearance attributes were considered as general and no timeperiod was specified. Please see the appended questionnaire where you can find detailed explanation on how the actual attributes and choices were explained to the patient. The text (p6) has been revised as follow: They were chosen to be general aspects of treatment that best described the alternative interventions on offer within the SINS trial i.e. surgery vs. imiquimod cream (Table 1). Please see revised text p16: A limitation is that participants in this study are unlikely to have experienced topical imiquimod, thus we could not assess how previous imiquimod experience influenced their preferences (including the risk of possible reactions/side effects to the treatment), whereas we could at least partly with surgical excision as some had already experienced it for a previous BCC. It would have been informative to analyse the DCE further,

4 7 The authors state they used a third status quo alternative although for me it is not clear why they used a fixed status quo alternative of which the levels are already included the other option alternative surgery. Could the authors provide more information about the reason for using this fixed status quo alternative because they also could have chosen (in this particular case) for a forced choice. Reviewer - Esther de Vries 1 The title does not seem to fully cover the content of the paper. The title suggests that the reader will get an answer to the question what are determinants of patients for treating BCC in general. The DCE only partly answers this question, only for the choice between sugery and imiquimod and of course only for the included attributes (although these will be the most important ones). 2 The abstract could be more informative, including most important conclusions on the most important determinants of patient preference. Similar to my first point: Results section of the abstract mainly gives the conclusion that patients prefer imiquimod, but in the abstract nothing can be read on the determinants of this patient preference. for example, comparing any changes in patient preferences later in the study after experiencing surgery or imiquimod cream. Also patients with both nodular and superficial BCC were included in the current analysis. Future studies could aim to collect larger sample sizes to perform a subgroup analysis according to the type of BCC affecting the patients. Many thanks for the comment. Multiple-choice designs (with the introduction of a third opt-out or status quo comparator) are extensively used in the literature. There is evidence that by not including an optout or status quo alternative when this reflects the reality, participation responses may be overestimated. As the status quo alternative can be either fixed or change across respondents, discussion with the experts informed the actual format of the status quo alternative. A fixed current situation was available to patients as standard practice in UK (see page 7). This was a particular case of surgery, and therefore its levels are included in the more general surgery alternative, where different combinations of surgery can be presented to the patients alongside the fixed current (surgery) alternative. Please see revised title: What determines patient preferences for treating basal cell carcinoma when comparing surgery vs imiquimod treatments? A discrete choice experiment survey from the SINS trial See revised results and conclusion, p 2 Results:.. Respondents were more likely to be worried about their cosmetic outcomes and side effects they might experience over and above their chance of clearance and cost.. Conclusion:.. Patients with BCC valued more imiquimod cream than alternative surgery options, and all

5 3 The attribute levels were presumably chosen to vary in order to be able to tell which aspects of which attributes influence patient choice. For the reader, it would be useful to also be informed on the levels which are considered most realistic for both treatment options. treatment characteristics were important for their choice of care. Many thanks for your useful comment; whilst the focus of this evaluation is to investigate respondent preferences (see preferred attribute levels derived from the regression output), the more realistic options can be found in their current experience (see their current surgery option). Re imiquimod option, participants in this study are unlikely to have experienced topical imiquimod, and therefore their realistic option should be informed by the trial output. See questionnaire in attachment. 4 As the authors describe, many respondents had difficulty in answering the questions. Did the authors consider to include a dominant choice set to test for rationality (i.e., a choice set which includes one treatment profile characterised by logically preferable levels on all attributes)? We are aware that one of the most common strategies for testing validity is the dominance test, although this type of test has been criticised for being too easy to pass (Ryan M, Bate A, Eastmond CJ, et al 2001) and it might not be applicable within a labelled exercise, where respondents could attach to each labelled service aspects of care beyond the ones described by the attributes. We therefore decided to test the theoretical validity of the responses looking at the direction and significance of the model attributes. Overall, all characteristics had the expected directions and confirmed the theoretical validity of the findings (see page 10).