2 | 19 Search strategies Papers were searched for in the PubMed, PsychInfo, Web of Science, Embase, and CINAHL databases. The search strategy included synonyms of the following concepts: ‘timely diagnosis’, ‘dementia’, ‘preferences,’ and ‘population’. MeSH terms, free text words, and equivalent index terms were used. The search was limited to English and Dutch language peer-reviewed journals published from January 2010 onwards. The time restriction was set to capture the most relevant considerations and preferences given the rapid developments in dementia diagnostics such as the possibility for biomarker-based diagnosis (28). Additionally, references of included studies were hand-searched. To optimize the search sensitivity and in line with previous studies that examine healthrelated preferences (29) and systematic search strategies for the construct preferences (30, 31), a broad definition of ‘preferences’ was used to determine eligibility for inclusion: “Patient perspectives, beliefs, expectations, goals, and the processes that individuals use in considering the potential benefits, harms, costs, and inconveniences of options in relation to each other” (32). We aimed at data triangulation on patient preferences by including GPs views on their patients’ preferences in our search strategy. The decision to start a diagnostic trajectory is usually made in general practice and GPs have longstanding relationships with their patients and knowledge of their patient’s personal life and their preferences (33). The full electronic search strategy for PubMed can be found in the appendix. The search strategies in the other databases were similar with equivalent index terms. The literature search was conducted in May 2020 and was updated in January and October 2021. Study selection Papers were considered for inclusion if they provided data regarding patients’ and SOs’(i.e. people that are directly involved in the decision-making process, such as spouses, children, other family members or friends) preferences for a dementia diagnosis and (shared) decision-making in that regard (Table 1). After removing duplicate papers, two researchers (IL and MH) independently excluded papers based on title and abstract. Next, the remaining papers were read full text. After each step, the researchers compared results and discussed differences. In cases of disagreement, a third researcher (CW) was consulted.
RkJQdWJsaXNoZXIy MTk4NDMw