Mariken Stegmann

We have not applied exclusion criteria. For all included studies, we extracted the following  characteristics: design and aim, setting, characteristics of participants, the primary  outcomes, how OPT outcomes were prioritised, and information about experiences.  The search, performed on November 6 2018, yielded four manuscripts describing three  studies, 15–18  and a protocol article 19  which we will not describe. Two authors (MES and SF)  assessed the quality of the studies as moderate using the GRADE‐system. 20  Characteristics of  all studies are shown in Table 1.   Case et al. performed a mixed‐methods cross‐sectional study of 356 persons aged 65 and  older living at home and compared the OPT with two other tools: the ‘Now vs Later’ tool and  the ‘Attitude scale’. 17  On the ‘Now vs later tool’, participants made a trade‐off between  quality of life at present or in the future. Using the ‘Attitude scale’, participants were asked if  they agreed with statements regarding different health outcomes (e.g. enduring side effects  now versus a better quality of life in the future). Fifty‐two percent of the patients thought  the OPT to be easy to use. When comparing the OPT to the other tools, most patients  preferred the OPT (41%), although no specific reason was mentioned. In this study, 21% of  the patients expected that the use of the OPT would change their care.   Ramer et al. performed a quantitative cross‐sectional study amongst 271 outpatients aged  ≥60 years with advanced non‐dialysis‐dependent chronic kidney disease to examine the  associations between outcome priorities, self‐reported health status, and acceptance of  end‐of‐life scenarios (e.g. kidney dialysis, cardiopulmonary resuscitation). No information  about patient experiences was provided. 18 Van Summeren et al. performed a mixed methods non‐controlled intervention study among  58 patients, aged ≥69 years, who used a large number of medications (polypharmacy). 15,16 General practitioners (GPs) and patients discussed the medication list taking the OPT score  into account, aiming for congruency between the indication, benefit and harm of the  different medications and the patient’s preferences. 15  In this study, 73‐92% of the patients  thought the OPT was understandable. 16 Considering healthcare providers, in the study by van Summeren GPs mentioned that the  goals ‘maintaining independence’ and ‘reducing other symptoms’ were the most difficult to  explain. 16  They also perceived the OPT as time consuming; about half an hour was needed to  elicit preferences, although less time was needed once experience was gained. However,  GPs indicated that the OPT did result in a better understanding of the patient’s life. Ramer  showed that, without the use of the OPT, healthcare providers have limited awareness of  the patients’ most important health outcomes. Providers’ perceptions what they considered  patients’ most important goals were correct in 35%, with an expected agreement of 25%,  resulting in a weighted Cohen’s k for patient‐provider concordance of 0.10 (95% CI ‐0.21 to  0.40). 18 The studies by van Summeren and by Ramer 15,18  described OPT scores on the four health  outcomes. In both studies, maintaining independence was most often prioritised as most  important outcome (35‐49% of the patients), followed by extending life (31‐35%), reducing  pain (9‐11%) and reducing other symptoms (6‐18%).  34 Chapter 3