Daan Hulsmans

148 Chapter 6 setting – done with the intention to document the client’s case file and keep colleagues up to date. Hence, care professionals received no instructions as to how extensive or comprehensive their reports should be. This meant that when a specific code was not identified from the records on a specific day, it may either have not been observed by care professional(s) or simply not been registered. Seemingly trivial happenings, such as giving complements will likely have occurred more often than that the coders coded in the records. Third, despite a-priori anonymization of the records, it was evident that the records included reports of many different (approximately > 30 different) care professionals. The richness of the described daily events likely partially depended on who reported and how much time that person had. Fourth, our three-step procedure was subject to many researcher’s degrees of freedom. The 11 staff-hypothesized (sub)themes that the participant’s clinician selected out of the thematic map, for example, remained a personal choice. Furthermore, the criterion we used to evaluate a threshold for instability (one tailed z-test at p < 0.05) is based on convention (cf. Olthof et al., 2020b; Schiepek et al., 2016; Wichers et al., 2018), but ultimately still a choice. On the other hand, there are no established guidelines available for a complex systems guided case study. This study also had strengths. First, by shedding light on events in the environmental that may ‘push’ the system into another state, our study adds to the (complex systems) psychological literature that has so far predominantly focused on instability preceding transitions (Lutz et al., 2012; Olthof et al., 2020b; Schreuder et al., 2020; Scott et al., 2017). Qualitative analyses of case records allowed us to distinguish everyday- from extraordinary events. Because this distinction was informant-based and not self-reported, it is possible that meaningful events were missed (here or in any step of our analysis). Future qualitative or mixed-methods research should further explore the nature of events that the individual perceives to ‘kickstart’ transitions. A second strength is that our research gives a helicopter view of day-by-day processes across several months. The majority of EMA research in borderline personality disorder studies within-day fluctuations. For our participant behavior did not only fluctuate within-days, also across timeperiods of multiple weeks or months. This may inspire EMA research in borderline personality disorder to consider further exploring fluctuations on slower timescales. Nevertheless, within-day processes remain relevant. Complex systems, after all, are characterized by interacting processes across many timescales (Olthof et al., 2023; Wallot & Kelty-Stephen, 2018). In our case, unobserved instability at shorter timescales (e.g., hour-to-hour) could

RkJQdWJsaXNoZXIy MTk4NDMw