How Does Task Loading in PICU Impact on Clinician Situational Awareness and Awareness of the Passage of Time?

Overview

There is little published literature on the subject of clinician perception of trends in patient observations on PICUs (Paediatric Intensive Care Units) and how this situational awareness is impacted on by clinical tasks. Human factors training cautions against the loss of situational awareness and time perception but there is no supporting observational data in the IC (Paediatric Intensive Care) setting. Perceptual loading theory hypothesises that, under low load situations, awareness extends to environmental features not directly related to the task at hand. However in high load situations awareness is restricted to the object of focused attention. Individuals experience different load from a given task depending on their skills and experience. Our pilot project intends to examine how clinician perception is affected by task loading. In our protocol two tasks are undertaken. The administrative task involves requesting a list of investigations from a clinical guideline. The technical task is the uncomplicated insertion of a central venous catheter (CVC) into a simulated vein. In each case the investigators will record proxies for perception: awareness of the passage of time, the time point at which monitoring changes are noted and retrospective recall of observation trends. Our protocol was designed with psychology input from Prof. Nilli Lavie, group leader of the UCL Attention \& Cognitive Control laboratory. Passage of time is measured by the participant pressing a foot pedal linked to timing software at every perceived 10 second interval of elapsed time. A retrospective estimate of total elapsed time will also be recorded. After the second task, immediate verbal questions are asked followed by a written questionnaire which contains questions targeting the participant's awareness of monitoring changes. Analysis will involve paired t testing of the timing interval and observation trend data to see if task loading significantly impacts accuracy.

Recruitment of participants Participants will be provided with the information sheet appended and given the opportunity to ask any questions before consenting to the study. They will be asked to sign the appended consent form. Their consent forms will be in no way linked to the data collected such that participants could be identified from their data. Randomisation Block randomisation will be used to ensure even distribution of individuals at each grade of experience between the allocation groups. A block size of 4 will be used with 2 strata (grade= Consultant, Other). The randomisation list was determined by an online randomisation service. The list generated is included as an appendix. The study cannot be blinded at the time of data capture given the significant obvious differences between the tasks. Study procedure Experimental scenarios: Each participant will undertake the timing control task first and then both the administrative and technical tasks, the order of these determined by the randomisation table. Timing control The participant sits and is asked to push a foot pedal every 10 seconds with no reference to external timekeeping. This task continues for 6 minutes. Participants are instructed: "Please press the foot pedal every 10 seconds, trying to be as accurate as possible. The task will last for several minutes." After 6 minutes the participant will be asked for the time period they believe has elapsed between the first pedal press and the last. Administrative task: Request investigations for a new admission Equipment required: PIMS-TS investigations pack Computer signed in to the EPIC learning environment for a dummy patient Simulated bedside monitor Varies observations over time in line with preset programming determined by the study team Foot pedal and data recorder for time logging Setup All equipment laid out on a table. The tissue pad is attached to the mannequin leg and surgically draped appropriately. The ultrasound probe is already in a sterile sheath, lying on the sterile field. Instructions to subject You are the fellow/consultant looking after Sophie. Sophie is a 2 year old girl with PIMS-TS. She is intubated and is stable on minimal ventilation and peripheral noradrenaline of 0.1mcg/kg/min. She has just been admitted to the unit. This is Sophie's bedspace, all the equipment you see is part of the simulation. Your task is to request the admission investigations according to the sheet provided. The nurse looking after the patient has had to step out of the bedspace to prepare their infusions and have asked you to complete your task whilst keeping an eye on the patient in the bedspace. I am the nurse in the next bedspace, if at any point you want to make a clinical intervention please continue with the task and verbalise anything you would like me to do. As before, when I ask you to begin, please press the foot pedal every 10 seconds, trying to be as accurate as possible. The task will last for several minutes. Events during the task: If this task is the first task undertaken, then the bedside monitor will display the same values throughout, aside from 1 desaturation. Heart rate: 125 bpm, blood pressure: 90/50 (65), respiratory rate: 25 , oxygen saturations: 98%, end tidal CO2: 4.5 mmHg. The oxygen saturation measure will fall to 88% for 5 seconds at 30 seconds into the task, triggering an alarm. Normal values for age are: Heart rate: 80-140 bpm, respiration rate: 20-28 (APLS), 5th centile systolic: 65+(2 x 2)= 69, 5th centile MAP: 50+(2 x 2)=54. At 30 seconds into the task there will be a 10 second desaturation to 88% with spontaneous normalisation to 98%. If this is the second task undertaken then values on the monitor will change during the task. There are 2 monitor scenarios, 'A' and 'B'. Allocation between these is determined in the randomisation table (see appendix). They are identical apart from 'A' being a falling trend and 'B' a rising trend Scenario A The oxygen saturation measure will fall to 88% for 5 seconds at 30 seconds into the task, triggering an alarm. The blood pressure will fall from an initial systolic/diastolic (mean) of 90/50 (63) to 75/48 (55) at 4 minutes and then remain static at that level. There will be no alarm. Other parameters (heart rate, respiratory rate, oxygen saturations, end-tidal CO2 and temperature) will not vary. Scenario B The oxygen saturation measure will fall to 88% for 5 seconds at 30 seconds into the task, triggering an alarm. The blood pressure will rise from an initial systolic/diastolic (mean) of 75/48 (55) to 90/50 (63) at 4 minutes and then remain static at that level. There will be no alarm. Other parameters (heart rate, respiratory rate, oxygen saturations, end-tidal CO2 and temperature) will not vary. End point: 6 minutes or completion of the task. The participant will be asked for the time period they believe has elapsed between the first pedal press and the last. Technical task Inserting a CVC. Equipment required: Child mannequin, intubated CVC insertion mannequin US machine + gel + probe cover CVC set Syringes, flush Cleaning- chlorprep lollipops Suture, needle holders, scissors, sharps bin Dressings + biopatch Gloves (non-sterile) Drape 5 +10ml syringes Sterile field drape Pen and paper for documentation Simulated bedside monitor Varies observations over time in line with preset programming determined by the study team Foot pedal and data recorder for time logging Setup All equipment laid out on a table. The tissue pad is attached to the mannequin leg and surgically draped appropriately. The ultrasound probe is already in a sterile sheath, lying on the sterile field. Instructions to subject You are the fellow/consultant looking after Sophie. Sophie is a 2 year old girl with PIMS-TS who requires central access for inotropes. She is intubated and is stable on minimal ventilation and peripheral noradrenaline of 0.1mcg/kg/min. This is the Sophie's bedspace, all the equipment you see is part of the simulation You have been asked to insert a femoral CVC and have already prepared your equipment.. The line is not flushed. You are already wearing the appropriate sterile gown. Please only use steristrips and dressings, with no sutures, to secure the line to the mannequin. Please document the procedure on the paper provided when you have completed insertion of the line. I am the nurse in the next bedspace, if at any point you want to make a clinical intervention please continue with the task and verbalise anything you would like me to do. As before, from when I ask you to begin, please press the foot pedal every 10 seconds, trying to be as accurate as possible. The task will last for several minutes. Given that this is a simulated mannequin please take all the time you need to familiarise yourself with the ultrasound appearance of the vessels before telling me you are ready to begin. Events during the task: Same as for the administrative task. End point 6 minutes or completion of task documentation. At the end of the task, progress through the checkpoints below will be noted by the experimenter: Equipment preparation / line flushing Cleaning of the procedure site Initial ultrasound scan Needle manipulation Guide wire insertion Tract dilation Line insertion Line suturing Line dressing Documentation completed The participant will be asked for the time period they believe has elapsed between the first pedal press and the last. Post- task review Following completion of the second task the participant will immediately be asked the following questions: "Did you see anything of note on the monitor aside from the desaturation" - their yes/no answer is recorded. "If there were changes in the blood pressure did you notice a fall or a rise" - their rise/fall answer is recorded. Was the extent of the fall/rise in systolic pressure: 5, 10,15, 20 or 25 mmHg- their selection is recorded. Which of the tasks did you find more challenging? And why? They are then asked to describe their subjective experience- "how were those tasks for you?" They are then asked- "How would you describe your familiarity with EPIC order requesting?"- Very, somewhat, limited, none They are then provided with the questionnaire from the appendix. Measurement tools used For each experimental run there are 2 variables The task order- administrative or technical completed first The trend in monitored observations- increasing or decreasing blood pressure in the second task The following parameters are measured during both tasks: Timestamps of foot pedal use by the participant Retrospective recall of total elapsed time Time to complete task/progress at experiment end Question responses Following completion of the tasks the data collected can be analysed as below. Time interval analysis: Intervals between pedal presses are extracted for each scenario (timing control/administrative/technical tasks) Derive: min/max/mean/interquartile range/standard deviation for each case Comparison of the individual's retrospective perceived elapsed time with actual elapsed time. Reported as percentage error of actual elapsed time. This will allow the following results to be reported: Mean deviation for each individual from their baseline in each of the tasks. Change in retrospective perceived elapsed time compared to actual elapsed time. Distribution of timing in the control task for each individual. Paired t-test comparison to test if there are significant differences in time interval distribution for each individual between control, administrative and technical tasks. Overall proportion of participants showing significant deviation from their own baseline in each of the administrative and technical tasks Population limits/mean Comparison of the mean perceived time interval between subgroups By task order By time within task (split into thirds- 2 minutes each) By grade By number of CVCs previously inserted By first 25% vs last 25% of participants Monitor trend perception analysis: Each participant will provide question answers from immediately after the second task. This will allow the following results to be reported: Rationale The first question "Did you see anything of note on the monitor" is chosen to highlight inattentional blindness and is asked immediately to avoid any impact from the role of memory or priming from other questions asked. This will be reported as proportion who noted a change The subsequent questions are asked to elicit responses from those who observed the trend but are not sufficiently confident in their recall to report it. From the responses the following can be derived: Correct identification of the blood pressure trend- reported as proportion of participants correct Correct identification of no trends in other parameters - reported as proportion of participants correct Accuracy of magnitude recall The blood pressure either falls or rises between a systolic/diastolic (mean) of 70/48 (55) and 90/50 (65). From the position of the x on the linear blood pressure scales the investigators can derive: The accuracy of the initial and final estimates compared with the actual values- reported for each task as the population mean and standard deviation of the percentage errors compared with the actual value. The accuracy of the magnitude of the change, determined through comparing the difference in the estimated initial and final values compared with the actual 20mmHg systolic/ 10mmHg diastolic change for each task. The population mean and standard deviation of the percentage errors compared with the actual change will be reported for each task. The investigators hypothesised that there would be a priming effect following administration of the questionnaire, with participants more likely to attend to blood pressure trends if the questions were asked prior to the second task. The invesitgators will therefore only ask the questions after the second task. The investigators can then compare between the subgroups, split by which of the tasks was performed second to examine the impact of task type. i.e the questionnaire analysis will have half the n of the overall study For the above the investigators can compare subgroups using Fishers exact test. By task order By grade By number of CVCs previously inserted By first 25% vs last 25% of participants