132 Chapter 6 As customary in the realist approach, some changes were made to the protocol registered on PROSPERO (CRD42023446437), while conducting this realist review. The title and research questions were revised during the review and updated as the review parameters were shaped. As the review progressed, the team felt the need to involve an expert in realist methodology to ensure adherence to the realist principles and to refine our analytical approach. Initial programme theory development The aim of the initial programme theory was to describe a hypothesis on how the implementation of MLCC can be successful. Prior to this review, the first and third author conducted 41 interviews and 8 focus-groups with various stakeholders on the subject, leading to an unpublished dataset. This dataset was input for formulating an initial programme theory. In addition, the first author conducted an unstructured search for evidence, and the first 3 authors drew on their experience as midwives. The formulated initial programme theory in the form of ‘if-then statements’ was presented in a Dutch stakeholder meeting to which 33 individuals were invited, including obstetricians, hospital-based midwives, community midwives, professional associations, health insurance companies, and a client organisation. Stakeholders were selected and invited based on their relevant expertise and to ensure the necessary diversity and variation in maternity care models required for this review. During this meeting, the if-then statements were confirmed, refuted, or adjusted. The resulting statements can be seen in online supplemental material 1. Searching process The search was set up in collaboration with a medical information specialist (KAZ). Search terms, including synonyms, closely related words, and keywords, were used as index terms or free-text words: “continuity of care”, “midwife”, and “implementation”. To test search terms, a pilot search was performed using several databases. After refining the search strategy, the following bibliographic databases were searched comprehensively for published and unpublished studies through August 1st, 2023: Ovid/Medline, Embase.com, Clarivate Analytics/Web of Science Core Collection, EBSCO/CINAHL, and EBSCO/APA PsycINFO. Limitations on human studies were applied in all databases. Additional searches consisted of hand-searching bibliographies and reference lists of the included studies, systematic reviews, grey literature (Google Scholar), relevant websites, national and international reports and guidelines, books, dissertations, and theses. Citation tracking was used to crosscheck whether all relevant studies had been identified. Only studies from high-income countries, as classified by the World Bank, conducted within the last 10 years were included, as contexts change over time. The searches contained no methodological search filters or language restrictions.
RkJQdWJsaXNoZXIy MTk4NDMw