Aim: In order to produce empirical evidence on the relationship between high bed occupancy and its potential hazards, this study correlates bed occupancy rates with hospital patient safety and staff overload indicators.
Methods: Data from nine medium to large scale public hospitals, all pertaining to the Hospital Foundation of Minas Gerais, Brazil, were gathered for the period January 2007 to June 2011. Indicators were aggregated by month, resulting in 486 observations (54 months x 9 hospitals). Bivariate linear regressions were performed, aiming to estimate the effect of bed occupancy rates on each response variable (hospital infection rates, institutional mortality and sick leave incidence). In addition, to directly test the hypothesis that bed occupancy rates over 85% are unsafe, it was included in the models as a categorical instead of continuous variable, using 85% as the cut-off value.
Results: Bed occupancy rates showed an inverse correlation to mortality rates (b = -0.056; P < 0.001) and presented no significant correlation to the nosocomial infection rates (P = 0.512). High bed occupancy (>85%) was associated with a slight increment of short sick leaves, especially short leaves (<7 days) (+0.14%; P = 0.008).
Conclusions: The increase in hospital loading was unexpectedly associated with reduced institutional mortality and was not related to nosocomial infection incidences. High bed occupancy was associated to a slight increment of short sick leaves.