Implementing capnography into practice for respiratory monitoring during sedation is considered a high priority by leading authorities in Canada and internationally. The Canadian Anesthesiologists' Society's position statement recommends that capnography should be available wherever moderate or deep sedation is used. The Academy of Medical Royal Colleges (UK) Standard and Guidance report on Safe Sedation Practice for Healthcare Procedures noted that while capnography is not a mandated practice, providers should consider implementing capnography as a long term goal. These recommendations are in place because sedated patients who are not monitored with capnography have frequent undetected, and therefore untreated, respiratory depression. Of note, though, these guidelines do not provide specific recommendations for how capnography should be implemented for nurse-administered sedation. The aim of this study is to determine if smart alarm guided treatment of respiratory depression using the Integrated Pulmonary Index is an effective way to implement capnography during nurse-administered sedation. The primary outcome is the number of seconds in an alert condition state without an intervention being applied. The IPI is intended to reduce the cognitive burden of synthesizing multiple sources of physiological monitoring input and hence lowering the threshold for triggering intervention by clinicians to support respiration. The primary outcome directly measures this concept by quantifying the time taken for an alert to trigger an intervention.

Higher values of the primary outcome will result from either a problem state that should have triggered an intervention but did not, or an 'inappropriate' alert (i.e. an alert that was not important enough to warrant immediate intervention.)

A hybrid effectiveness-implementation design will be used for this study, incorporating a parallel cluster randomized trial. Capnography monitoring will be implemented as standard practice for all patients undergoing procedures with nurse-administered sedation in the four departments participating in the study, as per recommendations for procedural sedation in Canada. Nurses working in the four departments will be invited to choose to participate in the randomized controlled trial component of the study. If they choose to participate, they will be randomized to use capnography either with or without the Integrated Pulmonary Index (IPI) enabled.

There will be establishment and stability phases where the implementation of capnography (with or without IPI enabled) into practice commences and is optimized through an audit and feedback process. Data collection for the primary outcome will be undertaken during the evaluation period. This design was chosen for two main reasons. First, it allows for the establishment phase, which will provide the time required for the research team to meaningfully engage with staff at each department and recruit nurses to participate in the randomized controlled trial component of the study. Second, the stability period will involve sustained engagement with clinicians in the department, regardless of assignment, in order to optimize the implementation of capnography monitoring (with or without the IPI enabled) into practice using a facilitated audit and feedback process.

Sedation management and monitoring

Medication

There will be no restrictions on sedation dosing for this study. It is typical for a bolus dose of 25-50mcg of fentanyl followed by 1-2mg of midazolam to be administered approximately two to five minutes prior to procedure commencement. Further doses of 1mg midazolam and 25mcg fentanyl are usually administered at the discretion of the proceduralist to maintain sedation.

Supplemental oxygen

Supplemental oxygen is routinely applied for patients unless contraindicated by pre-existing conditions, such as severe chronic obstructive pulmonary disease. The oxygen flow rate will be adjusted as per direction from the proceduralist.

Physiological monitoring

As part of usual practice in these departments, all patients receive all aspects of standard physiological monitoring including electrocardiography for heart rate and rhythm, non-invasive blood pressure measurements taken at regular 5-10 minute intervals (or more frequently if deemed required by nursing or medical staff) and continuous pulse oximetry. Assessments of respiratory rate and depth, as well as level of consciousness, are taken at regular intervals.

Additionally in this study, the standalone Capnostream 35p monitor (Medtronic) will be used. It displays a capnography waveform as well as end-tidal CO2 and respiratory rate. A nasal cannula or facemask that allows oral and nasal sampling of exhaled air as well as the delivery of supplemental oxygen will be used (Medtronic). The nasal cannula or facemask will be connected to the capnography monitoring device (Capnostream 35P, Medtronic). Nurses who choose to participate in the randomized controlled trial component of this hybrid effectiveness-implementation study, will be randomized to use capnography with or without the Integrated Pulmonary Index enabled. The IPI is a feature, which can be enabled in the Capnostream 35p devices. Nurses who are *not* participating in the randomized controlled trial component of the study will use capnography with or without the IPI feature enabled, at their own discretion. The Capnostream 35p monitor also collects SpO2 and heart rate measurements via pulse oximetry using a finger probe.

Establishment period

The establishment and stability periods will be guided by recommendations for improving the quality of clinical alarms from The Society for Critical Care Medicine Alarm and Alert Fatigue Task Force.

During the establishment period, each department will be provided with 2 Capnostream 35p monitors so that capnography monitoring can become part of routine practice for nurse-administered sedation. At this time, the research team will assist with providing education to nursing staff for capnography monitoring and the Integrated Pulmonary Index. The principal investigator will provide an overview of the Capnostream 35p device and how to use it, including the functionality of the Integrated Pulmonary Index. All nursing staff will be provided with printed information about how to access a standardized education program designed by Medtronic, which is certified for accreditation with the American Nurses Credentialing Centre and American Association for Respiratory Care (https://www.medtronic.com/covidien/en-us/clinical-education/catalog/healthstream-capnography-standard-of-care-for-procedural-sedation-monitoring.html). Specifically, in this course nurses will have access to education material covering topics including how to detect common capnographic waveform abnormalities, which indicate sedation-induced alveolar hypoventilation, and implement an appropriate intervention if indicated according to the alert conditions.

Stability period

The goal of the stability period is to optimize the alarm conditions set for the Capnostream 35p devices prior to the evaluation period. We will undertake an audit and feedback process for nurses choosing to participate in the randomized controlled trial component of the study. The audit and feedback process will evaluate the performance characteristics of the alerts triggered by the Capnostream 35p monitor (with or without IPI enabled depending on randomization). Alert performance characteristics to be audited include:

• Alarm burden (total number of alarms triggered by the Capnostream 35p).
• Number of appropriate alarms (defined as an alarm that triggered an intervention).
• Number of inappropriate alarms (defined as alarms that were triggered but manually dismissed by silencing the alarm).
• Duration of alert conditions (defined as the total time that an alert condition was active).

The Research Assistant will need to be present during procedures to observe and record the type of interventions implemented in response to capnography monitor alarms, as well as the time that the intervention was observed. A web-based data collection tool has been designed for this purpose (more detail is provided in the data collection section below).

The feedback of data collected during the audits for nurses who choose to participate in the randomized controlled trial will take an aggregated summary form per department. It will further delineate alarm performance characteristics depending on whether or not the IPI was enabled. The feedback will be presented on a webpage created for each department so that it can be updated approximately weekly as new audits are performed. The link to the webpage will be provided to nurses at the time of randomization. Access to the webpage will be restricted to users through an email address with an organizational domain (i.e. uhn.ca). The webpage will contain tabulated and other visual summaries of the audit data to highlight alarm performance characteristics, such as the number and duration of appropriate and inappropriate alarms, total alarms etc. In addition to the webpage, the Principal Investigator and/or Research Assistant will present summary findings from the audits at departmental meetings at least once per month to facilitate discussion amongst staff as to the appropriateness of current alarm settings and if changes are indicated.

The alarm conditions set for the Capnostream 35p monitor can be revised at any time throughout the study period at the discretion of the nurses based on insights from this audit of performance characteristics. This is an implementation study, employing an audit and feedback process to assist departmental staff to optimize capnography alarm settings (with or without the IPI enabled).

Evaluation period

Alarm performance for the Capnostream 35p monitor, now optimized for each capnography monitoring condition through the establishment and stability periods, will be compared between nurses randomized to use capnography alone or with the Integrated Pulmonary Index enabled.

Procedures

Participant selection and enrollment

Due to the hybrid effectiveness-implementation design to be used, all patients undergoing procedures in the departments included in this study will receive capnography monitoring. We anticipate that approximately 1200 patients will undergo procedures with capnography monitoring across the departments during the study overall. The target sample size for the evaluation period is 400 patient participants (sample size calculation presented in a section below).

This study involves only routinely collected clinical data and direct observation of nursing responses to capnography monitor alarms. As such, consent from patients will not be sought. The Tri-Council Policy Statement requires a study to fulfill conditions for waiver of consent, including that the alteration to consent requirement is necessary to address the research question. The integration of capnography monitoring with or without the IPI enabled is planned to be optimized in this implementation study through an audit and feedback process. As such, it will be impossible for individual patients to provide consent for participation in this study prior to the initiation of implementation and optimization of capnography into nurses routine care of sedated patients. In the 'evaluation' period specifically, a Research Assistant will observe procedures being performed with nurses who have chosen to participate in the randomized controlled trial component of the study on randomly selected days of the week until the target sample size is reached. The Research Assistant will be blinded to allocation status, as this is essential for studies that use clustered randomized trial designs to reduce risk of selection bias.

Data collection

This study involves only routinely collected clinical data and direct observation of nursing responses to capnography monitor alarms. During all phases ('establishment', 'stability' and 'evaluation'), we will retrospectively retrieve data recorded on the Capnostream 35p monitors using the USB port. Specifically, the 'Real-time Full Continuous Transfer' report will be downloaded. This report contains the real-time CO2 waveform plus real-time values of parameters (EtCO2, SpO2, pulse rate, respiratory rate, and IPI) with 1-second resolution by repeating numerical values for every 50ms along with the waveform. Other fields include the patient type (adult), report generation date and time, alarm occurrences, equipment advisory messages (eg. waveform not available, battery low, etc) and events entered into the monitor by a clinician (medication administrations etc).

During the 'stability' and 'evaluation' periods, a Research Assistant will be present during procedures to observe and record the type of interventions used, as well as the time that it was observed. A web-based data collection tool has been designed for this purpose, which will be accessed using an iPad. The times will be synced with the capnography monitoring device (Capnostream 35p).

During the 'stability and 'evaluation' periods, standard demographic and clinical characteristics will be collected by a Research Assistant. Procedural characteristics (type and duration) and sedation characteristics (medication type and dosages) will be collected for the purposes of the audit and feedback process in the 'stability' periods and for the purposes of describing the participant characteristics in the evaluation period.

Inclusion Criteria

Adult patients who are scheduled to undergo elective procedures with sedation administered by the nursing staff will be eligible for inclusion in the study.

Exclusion Criteria

Procedures performed without sedation or capnogrpahy monitoring

Study data, participant-level data sets (without identifying information) and code for statistical analyses will be shared in a publicly accessible repository at the time of publication.

With publication of results

Available through a public repository