Human resource allocation at intensive care unit (ICU) is essential for safety and quality in patient care. Although planning and organization are among the most effective instruments for health care management, human resource allocation varies across settings and is usually based on expert opinion, rather than on objective grounds. The identification of predictive factors for the length of physiotherapy session at the ICU could help with planning and management in health care services.

To identify predictive factors for the length of physiotherapy session applied to adult ICU patients.

This was a longitudinal panel study. The primary outcome was the physiotherapy session length, which was collected at one time point, however, the same patient could have more than one session length collected in different observation time points. Data were collected from a 12-bed adult ICU at a teaching, secondary-care public hospital, where common practice relies on the physiotherapists’ professional judgement to decide the time allotted and procedures used to manage each patient. All physiotherapy sessions applied to clinical and surgical patients were included. Sessions abruptly discontinued were excluded. A researcher followed one physiotherapist at a time during their entire work shift, measuring the duration of each physiotherapy session using a stopwatch. Physiotherapists signed informed consent form and provided information regarding their age and experience. Patient clinical and demographic data were collected from medical records. The study hypothesis was tested based on the patient and physiotherapist-related factors and the session length using a Multilevel Mixed Model. Sample size was estimated as 308 physiotherapy sessions (20 observations/predictor + 10%). The level of significance was p=0.05. Analysis was performed with software Jamovi 1.6.

The study assessed 339 physiotherapy sessions during 79 periods of observation, involving 181 patients and 19 physiotherapists. Average (SD) session length was 31.5 (14.5) minutes. The median number of patients assisted per physiotherapist per 6-hour shift was 5 (IQR: 4 to 5). Physiotherapists’ median age was 35 (26 to 39) years-old and ICU experience was 13.0 (0.4 to 16.0) years. Patients were mostly elder, post-surgery (38.7%), with current ICU length of stay of 5 (2 to 9) days. The Multilevel Mixed Model adjusted for outliers showed that current ICU length of stay [Estimate = 0.154 (0.027 to 0.281)], contraindication for out-of-bed mobilization [Estimate = -7.835 (-10.879 to -4.791)] and current use of sedatives, invasive mechanical ventilation, or vasoactive drugs [Estimate = 3.178 (0.223 to 6.133)] were associated with the length of physiotherapy session.

This was a single-center study; therefore, generalization should be made with caution. In our sample, factors related to the physiotherapist, such as age or experience, were not associated with session length. Contraindication for out-of-bed mobilization decreased session length while ICU length of stay and current use of sedatives, invasive mechanical ventilation or vasoactive drugs increased session length.

The identification of predictive factors for session length may help to estimate the number of patients that one physiotherapist is able to assist during the work shift, thus improving human resource allocation.