Search for a command to run...
Burnout among staff providing residential dementia care can undermine workforce sustainability and care quality. Evidence from Hungarian long-term residential social care settings remains limited, particularly regarding the combined role of multidimensional job demands and coping-related resource constraints. A county-level cross-sectional questionnaire survey was conducted among nurses and caregivers working in residential social institutions providing dementia care in Hungary. Data collection took place between 1 June and 31 July 2024 across 77 institutions. The final sample included 1,254 participants (ordinal regression sample: 1,207). Caregiving difficulties (13 items) and dementia-specific symptom burden (12 items) were assessed on 5-point scales; internal consistency was evaluated using Cronbach’s alpha. Principal component analysis was used to develop components of caregiving difficulty. Burnout severity was operationalized as a 4-level ordinal variable outcome (0–3), derived from participants’ responses regarding their own and others’ burnout. Predictors were examined using Proportional Odds Ordinal Logistic Regression (POLR) and a multilevel Cumulative Link Mixed Model (CLMM) accounting for institutional clustering. Caregiving difficulties showed excellent reliability (α = 0.928) and yielded three components explaining 73.19% of variance. Symptom burden reliability was also excellent (α = 0.949). Any burnout-related signal (levels 1–3) was reported by 46.4% of participants, while higher severity (levels 2–3) accounted for 19.3%. In adjusted ordinal models, higher subjective workload was associated with increased burnout severity (OR ≈ 1.46–1.53 across models, p < 0.001), as was more frequent care-related burden (OR ≈ 1.21–1.31, p < 0.01). Coping constraints were strongly related to higher burnout severity: reporting “no possibility to cope” was associated with increased odds (OR ≈ 2.86–2.91, p < 0.01). Self-reported prior learning about burnout was strongly negatively associated with burnout severity (OR ≈ 0.06–0.08, p < 0.001). Rest/relax frequency was not a significant predictor (OR = 1.09, 95% CI 0.95–1.24; p = 0.212). In the studied county, the level of burnout among professionals was primarily related to job expectations and limited coping options, while self-reported prior learning about burnout emerged as a potential protective factor. Organisational strategies that reduce excessive demands and expand access to coping resources—supported by targeted training—may help mitigate burnout risk in this setting.