Page 107 - Shared Guideline Development Experiences in Fertility Care

Online tool for patient partnership in guidelines
Considering the possibility of missing data, the N was enlarged to 200 patients. Based on the results of the cohort study of Brandes and colleagues, the hospital group was distributed between patients in an IVF clinic and a satellite clinic into a 2:3 ratio [31]. Accordingly, we used background characteristics of 80 patients visiting the regional IVF clinic and 120 female infertile patients visiting the regional satellite clinic. Both clinics were based in the east region of the Netherlands. Data of the hospital group were collected from patients’ electronic and written health records. Patients’ education level was derived from their occupations, since education level was not registered. If their occupation did not refer to the education level directly, data were recorded as missing.
Bene ts of the tool
Recommendations
To collect data on newly developed and modi ed recommendations, the following
aspects were automatically generated on and collected from the website: (1)
the total number of unique patients’ recommendations, (2) the total number of modi cations to these recommendations, (3) the total number of prioritizations on
these recommendations, (4) the change over time of the top ve recommendations 5 and (5) total number of performed prioritizations. Next, we collected all nal
top 3 or top 5 rankings per care phase and compared them with the initial 21 recommendations used to populate the tool at the start of the project. Additionally, we identi ed di erences in the top ves of recommendations per care phase over time.
Data analysis
All data collected from the website (participants’ background characteristics, number of new and modi ed recommendations, number of prioritizations) and scores on the items of the SUS were descriptively analysed using SPSS 16.0.
For calculating the SUS score, we rstly summed the score contributions of each item (0–4). For items 1, 3, 5, 7, and 9, the score contribution was calculated as the scale position minus 1. For items 2, 4, 6, 8, and 10, the contribution is 5 minus the scale position. Finally, the sum of the scores was multiplied by 2.5 to obtain the overall value of System Usability ranging from 0–100 [27–30]. For comparisons between users’ characteristics and the characteristics of the representative Dutch hospital population, we statistically analysed the data using the unpaired t-test (i.e.
105