Data collection
For hosting the internet survey and data collection, we used Sawtooth Software’s (Orem, UT) SSI Web. As recommended by earlier studies, the identification and selection of the final attributes and levels was based on literature review, qualitative research and an expert panel (19) (20) (21). The four stages as described by Heller et al were applied although we slightly customized these as seen in Figure 1, which means that we used several data sources in stage 1. To find out which attributes were important for DE patients, we firstly (Stage 1) collected data by performing a literature- and qualitative study. We performed a survey among 28 patients, one focus group with eight patients prior to their decision making (22) and a focus group with 10 gynaecologists with expertise on deep endometriosis. The results combined from stage 1 resulted in 158 attributes.
The second step (Stage 2) in attribute development is data reduction. This was achieved by frequency and rank order by the researchers (JeM and JS) combined with thematic analysis (grouping attributes with more or less the same theme), which resulted in 28 attributes. In the third and fourth stage, concerning respectively removing inappropriate attributes and wording of attributes, we selected attributes which comply with the research question, are relevant for the DE population, and are intelligible for all patients (23). The last process of attribute selection was performed in collaboration with all members of the research group, which resulted in eight final attributes (Table 1).
The first part of the survey included questions about baseline characteristics (i.e. age, educational level, medication use, medical history and pain symptoms), surgical fear measured with the validated Dutch surgical fear questionnaire (SFQ) (24) and three health literacy screening questions. We included the short surgical fear questionnaire, while we hypothesized that the fear for surgery could influence the results of the DCE. Women with surgical fear could have a tendency towards choosing the conservative treatment compared to women with less surgical fear. The surgical fear has a subscale sum score for short -and long term fear that ranges from 0 (no fear) to 40 (very afraid), the total sum score ranges from 0 (no fear) to 80 (very afraid). Pain was recorded on a numeric rating scale, patients rated their pain intensity (0 = no pain and 10 = maximum pain, or inapplicable option). The pain intensity score was obtained for menstrual cycle and non-cycle related pain, on the dimensions of dysmenorrhea, pelvic pain, dyspareunia, dysuria and dyschezia.
Part two of this survey included information about the DCE (Supplementary table S2 includes the attributes explanation). Prior to the DCE, we asked the women to rank the eight attributes from most important (1) to least important (8) when making a treatment decision. In the ranking we did not distinguish between conservative or surgical treatment. To get familiar with the concept of a DCE, a simple DCE question for choosing a phone. Subsequently, the actual DCE was presented with 10 choice sets. Each choice set consisted of two hypothetical treatment options labeled as pharmaceutical (conservative) and surgical treatment. The reason for choosing a labeled DCE is that both conservative and surgical treatment have a number of specific attributes and levels that are not generic for both treatments. The women were asked to choose their preferred treatment of choice for each of the 10 choice sets (Figure 2). The conservative treatment option included specific attributes like side effects of hormonal treatment: depressive feelings and the chance of developing osteoporosis. The surgical option included treatment specific risks like, the chance of getting a temporary stoma and the chance of permanent intestinal symptoms (LARS) ( Table 1).