The purpose of our study was to investigate which organizational levels and factors determine the number of resident handlings in eldercare. We conducted a multi-level study, stratified on day and evening shifts, including information on four levels: nursing homes (n = 20), wards within nursing homes (day, n = 120; evening, n = 107), eldercare workers within wards (day, n = 619; evening, n = 382), and within eldercare workers (i.e., days within eldercare workers; day, n = 5572; evening, n = 2373). We evaluated the influence of each level on the number of resident handlings using variance components analysis and multivariate generalized linear mixed models. All four levels contributed to the total variance in resident handlings during day and evening shifts, with 13%/20% at “nursing homes”, 21%/33% at “wards within nursing homes”, 25%/31% at “elder-care workers within wards”, and 41%/16% “within eldercare workers”, respectively. The percentage of residents with a higher need for physical assistance, number of residents per shift, occupational position (only within day shifts), and working hours per week (only within day shifts) were significantly associated with the number of resident handlings performed per shift. Interventions aiming to modify number of resident handlings in eldercare ought to target all levels of the eldercare organization.
BAYESIAN INFERENCE ON VARIANCE COMPONENTS IN GENERALIZED LINEAR MIXED MODELS: AN EVALUATION OF DIFFERENT METHODS
