Prevalence, incidence and associated factors of pressure injuries in hospices: A multicentre prospective longitudinal study
Marco Artico,Michela Piredda,Daniela D'Angelo,Maura Lusignani,Diana Giannarelli,Anna Marchetti,Cosimo De Chirico,Chiara Mastroianni,Maria Grazia De Marinis
DOI: https://doi.org/10.1016/j.ijnurstu.2020.103760
Abstract:Background: Patients in palliative care are the population cohort that most frequently develop pressure injuries, severely impacting their quality of life. Data from prospective studies on the prevalence and incidence of pressure injuries in hospices are lacking. Aim: To describe the point prevalence and cumulative incidence of pressure injuries in patients admitted to residential hospices, and to analyze their predictive factors over time. Design: Multicentre prospective longitudinal observational study. Setting/participants: Adult patients (n = 992) enrolled in 13 Italian residential hospices, with a minimum sample of 280 for each macro-region (North, center, South/Islands). Methods: Assessments including the Karnofsky Performance Status, Braden, Edmonton Symptom Assessment System Revised scales and pressure injury staging according to National Pressure Ulcer Advisory Panel were conducted at least every four days, from admission to patients' death/discharge. Results: The 7,967 observations recorded provided prevalence and incidence rates of 34.1% and 26.5%, respectively. The logistic regression model showed non-cancer disease (OR = 2.39, 95%CI = 1.65-3.47), age >80 (OR = 2.01, 95%CI = 1.49-2.71), Braden score 'at risk' (OR = 1.92, 95%CI = 1.17-3.14), urinary catheter (OR = 1.96, 95%CI = 1.40-2.75), drowsiness (OR = 1.41, 95%CI = 1.02-1.95) and artificial nutrition (OR = 1.47, 95%CI = 1.01-2.14) as the variables associated with pressure injury at admission. The generalized estimating equations models, built on the timeframes for observation groups, revealed male gender (OR = 1.68, 95%CI = 1.01-2.79) and Braden score 'at risk' (OR = 4.45, 95%CI = 1.74-11.34) as predictive factors of a new pressure injury developed up to three weeks before a patient's death, while in the last ten days of life these predictors were replaced by diagnosis of cancer (OR = 1.80, 95%CI = 1.11-2.91), worsening pain (OR = 1.65, 95%CI = 1.10-2.49), drowsiness (OR = 1.79, 95%CI = 1.25-2.57) and dyspnea (OR = 1.48, 95%CI = 1.01-2.18). Conclusions: The high incidence and prevalence of pressure injuries confirm the importance of palliative care nurses continuously focusing on prevention and management strategies. In the last three weeks of a patient's life, the predictive power of the Braden scale for a new pressure injury is not confirmed, throwing doubt on the effectiveness of aimed interventions at modifying risk factors. Along the different disease trajectories, pressure injuries developed during the instability/worsening phases of illness, occurring before hospice admission for non-cancer patients and in the end-of-life phase for cancer patients. Despite continuous provision of appropriate interventions, most of the new pressure injuries were detected during the last ten days of a patient's' life and assessed as 'unavoidable'. These results are crucial to guiding palliative care nursing plans during the different phases of illness, and to predicting care needs, possible management strategies ('wound management' vs. 'wound palliation'), and resource utilization.