Risk factors for early and late unplanned readmissions to the intensive care unit – results of a survey among intensivist

Abstract

Introduction. The knowledge and experience of anesthesiologists about predictors of unplanned ICU (intensive care unit) readmission are implemented into practical models for their prevention. However, differences in the time before patient’s ICU readmission make significant adjustments to the possibility of their implementation.The objective was to analyze intensivist opinions about predictors of early and late unplanned ICU readmissions. Materials and methods. The survey of intensivists was conducted in 9 hospitals in Saint Petersburg and the Leningrad Region from July 2023 to July 2024. Statistical information processing was carried out using the Jamovi software package.Results. The study included 381 questionnaires. Long ICU stay ( > 14 days) (AOR: 0.373; 95% CI: 0.183–0.758, p = 0.006; 0.492; 0.246–0.985, p = 0.045), emergency surgeries and procedural complications (1.283; 1.071–1.537, p = 0.007; 1.387; 1.136–1.694, p = 0.001), as well as lack of data on oxygen therapy and respiratory support in the transfer epicrisis (0.315; 0.172–0.576, p < 0.001; 0.505; 0.278–0.919, p = 0.025) increase the chances of 24- and 48-hour unplanned ICU readmission. Transfer of patients from the ICU to «off-hours» (1.244; 1.020–1.517, p = 0.031; 1.518; 1.243–1.853, p < 0.001) was a risk factor for the 48- and 72-hour periods. Registration of adverse acute cardiovascular events (2.876; 1.368–6.047, p = 0.005; 2.578; 1.390–4.780, p = 0.003) increased unplanned ICU readmission in the 48-hour and 7-day periods. Sepsis in a patient in the ICU was considered by respondents to be an independent predictor of ICU readmissions for all periods except the 24-hour period. Characteristic predictors of unplanned ICU readmissions also were: for 24-hour – the impossibility of conducting «round-the-clock» laboratory and instrumental diagnostics (0.764; 0.639–0.914, p = 0.003), patient readmission due to the need to free up a ICU bed (1.345; 1.138–1.589, p = 0.001), failure to hold council (0.507; 0.270–0.954, p = 0.035); for 72-hour – physician experience (0.968; 0.939–0.997, p = 0.033); for 14 days – Mechanical Ventilation > 7 days (1.674; 1.025–2.734, p = 0.040).Conclusion. The analysis of the opinions of intensivists made it possible to identify independent and determine modifiable predictors of early and late unplanned readmission of the patient to the ICU.

Keywords

• survey (questionnaire)
• intensivist
• unplanned readmission
• predictors
• intensive care unit (icu)