Ethics approval and consent to participate
We received ethics approval from the research ethics committee of the Hospital Sírio Libanês/ Sociedade Beneficiente de Senhoras (CEPesq/HSL), with a reference number: CEPesq HSL2019-84. This committee in addition to releasing the research protocol, waived Informed consent form since the study was a retrospective review ofdata from the medical records or hospital electronic database. The methods were carried out in accordance with the relevant guidelines and regulation.
The study was conducted at the Sírio-Libanês Hospital, São Paulo, Brazil. The multifaceted improvement strategy was evaluated using a before-after study (Figure 1). The data were collected based on the physiotherapy service indicator of the out-of-bed mobilization rate. Consecutive patients who meet the inclusion and exclusion criteria of the study were part of the sample. Therefore, only patients aged > 18 years who were included in the indicator screening, without a mobility contraindication were considered. Contraindication was defined as any hemodynamic, neurological, or respiratory instability, medical contraindications, medical indication to priorities comfort measures, patient refusal, and admission or discharge day from the unit.
The “before” period corresponded to the previous months to the improvement strategy implementation, while the “after” period to the time following the executed intervention. The data were obtained in a decreasing and growing way in the months until the sample size was achieved.
This study is a quality improvement project, and the revised Standards for Quality Improvement Reporting Excellence (SQUIRE2) guidelines were followed.
Figure 1.Study design
During the study period, the adult ICUs consisted of 49-beds. Two general ICUs (21 beds), one neurological ICU (8 beds), and two cardiologic ICUs (20 beds). The staff was composed of a physician (staff-to-patient ratio 1: 5), registered nurse (staff-to-patient ratio 1: 2), and physiotherapist (staff-to-patient ratio 1: 5). A daily multidisciplinary round to determine the goals of care was performed in both periods, considering the ABCDEF bundle.
Data were collected from admission to hospital discharge from medical records or hospital electronic databases. Patient baseline information including demographics, comorbidities, and severity of illness at ICU admission were obtained from Epimed Solutions ®.
The verticalization and non-compliance rate collection was made by the same professional in charge for the institutional indicator that used a standardized checklist, to maintain greater reliability and validity. We also considered variables related to risk factors as ICU-acquired weakness (ICU-AW) (mechanical ventilation > 72 hours, sedation, analgesia, neuromuscular blockers, corticosteroids, sepsis, septic shock, and immobilism, defined as the permanence on bed for more than 50% of the day), related to the outcomes (length of stay, MRC or SOMS, and barriers for mobilization), and the highest achieved mobility landmark of the day, those information’s were collected by the project researchers through our institutional electronic medical records. All perceived barriers were also recorded. If any patient’s information was missing, the researchers or the professional in charge of the institutional indicator could access the assistance team. The checklist and the spreadsheet used to compute de data were double-checked to avoid any data loss or incomplete data.
To elaborate the improvement strategy, initially, we performed a summary of the evidence considering out-of-bed mobilization. Posteriorly, we optioned to understand the problem in the perception of the multidisciplinary team through a brainstorming during a workshop freely listed by the participants, with results were grouped in an Ishikawa diagram, and an online multiple-choice questionnaire. Both addressing the modifiable barriers related to environment, patient, staff, and process. Finally, the team perception was paired with the data obtained from a meticulous verification of the patient’s medical records to elucidate modifiable barriers involved in the cases of patients who were able to be mobilized and were not24.
After analyzing the results of these activities, it was verified the importance of improving communication between the characters involved, planning, and individualizing the process considering the specific barriers at the moment, in addition to including the patient and family more actively. Considering these points, a visual tool was developed named as “mobility clock” (Figure 2) to simultaneously quantifies, informs, and monitors the patient's functional level. This, instead of hours, it displays the different landmarks of mobility based on the ICU mobility scale25.
An action plan was prepared and set in motion to inaugurate the mobility clock in the week of mobilization. To sensitize the team about the importance of early mobilization, the weak was opened with a talk show where a patient who developed quadriplegic and his family members, in addition to the professionals involved in the care, shared their experience during the period of hospitalization in the ICU and the impact of this, after hospital discharge. During this week, a lecture was given with updated data from the literature on out-of-bed mobilization in the ICU, demonstrating at the end the result of the internal audit of our modifiable barriers and presenting the “mobility clock” through an animation produced specifically to explain its development and application. At the end a challenge that lasted four weeks, was opened, to formalize and establish its use in a routine way by the team. During this period a podcast explaining how to use the “mobility clock” and the challenge rules was broadcasted on the institution’s channels. Banners referring to mobility were displayed at the entrance of each unit and to motivate the staff to the cause stickersand chocolates were distributed 25.
The “mobility clock” was fixed in the ICU rooms to be visible to the patient, family, and staff.Instead of hours, it displays the different landmarks of mobility. One clock pointer marks the mobility level that was set as the goal and the other the landmark that was performed by the patient. To improve the communication in the multidisciplinary round, the expected goal per patient was determined considering the barriers presented, as well as the physical condition of each individual. At this time, if possible, the importance of reaching the chosen mobility landmark on the day was explained to the patient and family. The clock hand that corresponded to the landmark achieved by the patient was moved during the day as soon as the mobility level was reached.
During the intervention period, no other institutional strategy regarding early mobilization was employed.
Figure 2.The mobility clock monitors the level of mobility in the intensive care units of Hospital Sírio-Libanês and is based on the ICU mobility scale. It presents ten mobility milestones (the higher the score, the higher the mobility level achieved by the patient). One of the hands of the clock represents the mobility level planned by the multidisciplinary team for the patient during the shift (goal), and the other, represents what was achieved. In the example above, the objective elaborated by the team was to “march on spot” (level 6) and the milestone achieved was to “sit on the edge of bed” (level 3). Thus, the objective was not reached because the level of mobility achieved was lower than planned.
The sample size (at least 88 patients per period) was calculated based on a previous pilot study considering the number necessary to reduce the non-compliance with the protocol on 10%, given a two-tailed type 1 error of 5% and a power of 80%. Quantitative continuous variables were compared using the Mann–Whitney U test for non-normally distributed variables with interquartile values used to represent data dispersion. The means of normally distributed variables were compared using Student’s t-test. Pearson’s chi-squared test and Fisher’s exact test were used to compare the categorical variables. The significance level was set at p≤ 0.05 (two-tailed).