This protocol describes the DAILY-FLEX study, which investigates how daily work and caregiving demands and resources influence domain-specific well-being among employed family caregivers of people with dementia. The study team framed psychological flexibility (PF) as a momentary, volatile personal resource that may mediate bidirectional spillover between work and caregiving. Investigators reported specific aims: to quantify momentary and daily cross-domain spillover, to test lagged associations between contextual conditions and subsequent functioning, and to examine PF as a within-person mediator of these processes.
The protocol positions PF as a dynamic capacity rooted in Acceptance and Commitment Therapy (ACT), operationalized using the Hexaflex theoretical foundation and the Triflex domains—Open Up, Be Present, and Do What Matters—when measuring daily functioning. The investigators hypothesize that situational demands will reduce momentary PF and thereby increase negative spillover, whereas contextual resources will enhance PF and support positive spillover and maintained engagement.
DAILY-FLEX is an observational intensive longitudinal study combining baseline assessment with Experience Sampling Methodology (ESM). Participants complete a baseline questionnaire followed by a 14-day ESM burst. During those 14 days, the study uses eight semi-random momentary assessments per day plus a single evening questionnaire each evening. The m-Path smartphone application is specified as the delivery platform for all momentary and evening surveys.
This design captures within-day fluctuations and day-to-day variation in contextual demands, resources, PF, affect, and work engagement. The protocol emphasizes within-person inference: analyses will focus on temporal, lagged associations to explore how a momentary or daily state in one domain predicts later functioning in the other domain. The study team noted that recruitment and data collection are ongoing; this protocol does not report participant-level outcomes.
Momentary assessments collect proximal, context-sensitive indicators. The protocol lists contextual demands and resources relevant to either the work or caregiving environment, momentary PF, caregiving-related affect, and work engagement as the primary momentary items. Evening questionnaires aggregate daily experiences and include measures of daily work and caregiving demands/resources, daily PF, general well-being, and daily job satisfaction. The investigators aligned PF measurement with ACT process models, using the Triflex organization to map items onto Open Up, Be Present, and Do What Matters domains.
The protocol does not present item-level psychometric results from this study because data collection is ongoing. Investigators did, however, link the conceptual measures to theoretical constructs from the Work–Home Resources model, arguing that volatile personal resources like PF transmit the effects of contextual demands and resources to downstream functioning.
Planned analyses include multilevel mixed-effects models to account for nested momentary observations within days and days within individuals. The study team intends to examine lagged cross-domain spillover—testing, for example, whether a momentary work demand predicts subsequent caregiving affect or whether caregiving demands predict later work engagement. Longitudinal mediation analyses are specified to test PF as a within-person mediator in these temporal chains.
The protocol explicitly frames PF as a proximal mediator: situational demands or resources in one domain are expected to change momentary PF, which in turn is expected to influence subsequent functioning in the other domain. The authors noted that these analyses rely on within-person variation and time-lagged modelling to strengthen causal inferences, but the document does not report model estimates or statistical power calculations because the primary data collection remains in progress.
Investigators plan to recruit a minimum of 30 employed family caregivers of people with dementia in the Netherlands. Recruitment strategies mentioned include memory clinics, caregiver networks, and community outreach. The protocol clarifies that this article is a study protocol and does not report participant-level analyses; recruitment and data collection were ongoing at the time of publication.
The study received funding from Alzheimer Nederland, the Association for Contextual Behavioral Science BeNe Chapter, and the Adriana van Rinsen-Ponsen Foundation, and the authors declared no competing interests. Regarding data sharing, the protocol states that de-identified data and accompanying analysis code will be made publicly available on the Open Science Framework where permitted by ethical approval, participant consent, institutional policies, and applicable data-protection requirements. If ethical or legal restrictions prevent public release, the repository record will specify the institutional contact and procedure to request access.
The protocol acknowledges several inherent constraints. It does not report results or participant-level characteristics because recruitment and data collection are ongoing. The planned minimum sample size is specified, but statistical power for complex lagged mediation models is not presented in this protocol. Item-level psychometrics for the momentary PF operationalization in this sample are not yet available in the protocol document. The study team did not report any intervention or randomization procedure; the design is strictly observational, which limits causal inference despite the temporal sequencing of ESM data.
Open questions remain that the protocol leaves for the primary results publication: the achieved sample size, the distribution of work and caregiving contexts in the recruited cohort, observed effect sizes for lagged spillover paths, and the empirical behavior of the PF measures in momentary data. The investigators reported plans for public data deposition conditional on approvals, but final data release will depend on ethical clearance, participant consent, and institutional policies.