Tele–Intensive Care Unit Collaboration to Decrease Pressure Injuries: A Quality Improvement Project

In 2017, a nurse educator within the academic medical center surgical step-down unit initiated a practice change that required a 2-RN skin assessment for all patients newly admitted or transferred to the unit. This internal process change resulted in a 57% decrease in unit-acquired PIs. Unfortunately, the attempt to implement this best practice across the organization had variable results.

At the beginning of the COVID-19 pandemic, all ICUs at the academic medical center⁷  and many of the community hospitals within the 11-hospital health care system observed increases in hospital-acquired PIs. Skin changes related to COVID-19 are not pressure-related injuries yet may have been incorrectly identified as such, contributing to the increase in PI numbers. Due to rapidly changing events during the pandemic, an RN-led interprofessional team was formed to investigate how to reduce or prevent PIs amid worsening patient acuity and outcomes combined with restricted nurse time at the bedside. The pandemic fostered ingenuity and led to this novel approach for early identification, prevention, and treatment.

Nurse PI experts noted a decrease in documentation of PIs upon admission and a subsequent increase in hospital-acquired PI rates. Increased hospital-acquired PI rates were credited to a lack of documented 2-RN skin assessments within 24 hours of patient arrival to the ICU; skin injuries found during these assessments should have been categorized as present-on-admission PIs. The wound, ostomy, and continence RN (WOCN) team member described performing skin consultations with tele-ICU video technology to conserve limited personal protective equipment and reduce time spent in rooms of patients with COVID-19. The team chose to investigate the use of tele-ICU to assess patients’ skin upon ICU admission or transfer from another unit into the ICU.

Tele-ICU RNs provide a proactive and unique support system for ICU clinicians.¹²  The health system instituted the tele-ICU in 2016 at the academic medical center and has continued the growth of this resource to the present, allowing inclusion of all project ICUs. The technology platform in the tele-ICU allows nurses to use high-resolution video with the capability for video conferencing (Zoom Video Communications, Inc) and camera movement to assess patients for subtle changes for various reasons, including performing skin assessments. Thus, a partnership between tele-ICU and bedside ICU RNs could provide timely 2-RN skin assessments of critically ill patients to identify and document present-on-admission PIs. A process improvement project was undertaken to determine the success of this innovative approach.

The purpose of this project was to examine the impact of collaboration between tele-ICU RNs (using video technology) and bedside RNs to conduct a 2-RN skin assessment within 24 hours of admission for all ICU patients to identify and document present-on-admission PIs. The primary aim was to reduce unit-acquired PIs by at least 50% by accurately categorizing present-on-admission PIs to eliminate their incorrect categorization as hospital-acquired PIs and by promptly initiating prevention and treatment measures. A secondary aim was to achieve at least 90% overall compliance with 2-RN skin assessment documentation with the tele-ICU RNs within the first 24 hours of ICU admission.

Tele-ICU Collaboration to Decrease Pressure Injuries With Ostomy Wound and ICU Nurses (TIC DOWN-PI) was an interrupted time series quality improvement project to evaluate the change in hospital-acquired PIs attributed to proper identification of PIs on admission to the ICU and early prevention strategies. For the quality improvement analysis preimplementation period, crude data were obtained from January 2021 through September 2021 and excluded October 2021; thus, only 9 months of data were used before implementation. Before project implementation, WOCNs helped differentiate PIs from COVID-19 skin alterations. October, the transition month, was not included in the crude data because implementation started in the middle of the month. Data obtained from November 2021 through April 2023 were evaluated monthly to analyze trends in the postimplementation period. Due to the confounding factors of patients with COVID-19 admitted in 2021, the interrupted time series approach was used to account for a change in patient population. Monthly data collection included PIs that were classified as present on admission and documentation compliance for 2-RN skin assessments. The project was conducted in 1 academic and 2 community medical ICUs with a total of 79 ICU beds.

The interprofessional core team consisted of tele-ICU RNs, bedside ICU RNs, staff educators, unit-specific project champions, WOCNs, clinical nurse managers and directors, PI data experts, research scientists, a statistician, a business intelligence solution developer, a tele-ICU physician director, and a project manager. Wound, ostomy, and continence RNs were consulted using video technology for difficult-to-assess skin alterations. The WOCNs reported an increase in assessment ability due to the high-quality visualization of all skin tones that may not have been visible at the bedside. The TIC DOWN-PI project employed the best practice of a 2-RN skin assessment by (1) identifying PIs using video technology and tele-ICU RNs as second assessors within 24 hours of patient admission, (2) implementing prevention measures early using a best-practice checklist, and (3) providing documentation support from tele-ICU RNs.

When a patient was newly admitted to the ICU, tele-ICU RNs obtained report and coordinated with the bedside team to select a time to complete a collaborative head-to-toe skin assessment. The tele-ICU RNs announced their visual entry into the room and started using a high-definition video technology camera to examine high-risk body site locations simultaneously with bedside RNs. When the assessment was complete, the tele-ICU RNs reviewed a checklist and recommended individualized interventions (eg, low-air-loss bed, barrier creams, foam dressings on bony prominences, nutrition consultation, etc) to prevent the development of PIs. The bedside RN placed a WOCN consultation request to validate the skin assessment and assist with a treatment plan if stage 3, stage 4, deep tissue, or unstageable PIs or unclear skin alterations were identified upon admission.

Historically, tele-ICU RNs did not document their findings in the electronic medical record (EMR). The TIC DOWN-PI project leads established workflows, education, communication, and collaboration between tele-ICU and ICU bedside RNs to standardize processes for 2-RN skin assessment documentation in an attempt to reduce the documentation burden for the bedside RNs. The tele-ICU RNs documented the completed 2-RN skin assessment in the EMR. If a PI was identified, tele-ICU RNs would open the PI parameters and include locations, cause, and descriptions of the injuries after concurring with the bedside RNs. In addition to completing EMR documentation, tele-ICU RNs documented several variables in an online HIPAA (Health Insurance Portability and Accountability Act)–compliant database (Research Electronic Data Capture [RED-Cap], projectredcap.org). These variables included patient and unit demographics, collaborative 2-RN skin assessment completion, reasons for incomplete skin assessments, PI identification, prevention intervention discussions, and time spent for each patient admitted to the ICU.

Initially, project leaders spent 30 minutes to 3 hours per day cleaning data. They quickly recognized the lack of standardized reporting across ICUs, inefficient data extraction, absence of infographics, and delayed communication of compliance to leaders. The time commitment for data cleaning threatened sustainability. The tele-ICU data entry in REDCap was compared with the EMR for validity of 2-RN skin assessment compliance documentation. A business intelligence solution developer was consulted to assist with automatic cleaning, blending, and aggregating of EMR and REDCap data to reveal real-time feedback of process metrics. After consultation and new data report builds, selected project team members received daily, automatic emails from the previous 24 hours comparing ICU admissions and missed skin assessments as well as retrospective trend reports. Unit leaders were notified if skin assessments were not documented. Unit leaders then consulted with the patient’s assigned bedside RN to explore reasons for missed skin assessments and provide individualized education. Project leads validated the accuracy of the updated reports.

To show the impact of the interventions, daily automated reports that reflected documentation in the previous 24 hours were emailed to the ICU leaders. These customized reports were generated from EMR and RED-Cap data after project leaders reviewed missing data. The daily emailed reports displayed the total volume of admissions, 2-RN skin assessment compliance, and number of PIs identified.

This project posed no risk to patients other than potential breach of confidentiality. All data were stored in a password-protected file. All patients received current standards of care based on PI advisory panel guidelines.¹³  Only patients with imminent death were excluded. The institutional review board reviewed the TIC DOWN-PI proposal and deemed it a non–human subject research, quality improvement project.

The crude number of PIs classified as present on admission and the rate of 2-RN skin assessment documentation were calculated per ICU admissions each month overall and for the postimplementation period, during which tele-ICU RNs served as second assessors. The 2-RN skin assessment compliance was calculated by the following formula: number of intervention documentation entries (number of 2-RN skin assessment documented) divided by the number of opportunities (number of admissions). The percent compliance was calculated monthly and subsequently averaged per all months in the preimplementation and postimplementation time periods. The averaged percent compliance data were used to calculate percentages of change between the 2 periods.

Segmented regression analysis was used to estimate changes between preimplementation and postimplementation unit-acquired PIs, total PIs present on admission, and 2-RN skin assessment compliance over the course of the project. The preimplementation period for the segmented regression analysis included October 2021; hence, 10 months of data in the preimplementation period were analyzed. Significance was set at α = .05. Analyses were conducted using Stata/BE, release 17 (StataCorp LLC). There were no missing data to consider during the analysis.

Monitoring TIC DOWN-PI outcomes required comprehensive dashboards to show the success of the project over the extended period. With the business intelligence solution developer’s expertise, consolidated EMR and RED-Cap data were integrated into analytics software to generate a visual report to demonstrate real-time updates of process metrics compared with overall unit-acquired PI incidence rates. The report shown in Figure 1 demonstrates that as 2-RN skin assessments increased, unit-acquired PI incidence rates decreased compared with preimplementation rates. At the start of implementation in November 2021, 2-RN skin assessments increased. Beginning in April 2022, PI rates showed a sustained decrease.

For 4723 admissions, 2-RN skin assessment compliance increased from 46.9% during the 9-month preimplementation period to 80.8% during the 18-month postimplementation period, showing that compliance increased by 72.3% (Figure 2). Participation of tele-ICU RNs as second assessors was 63.2% overall and increased from 22% in November 2021 to 75% in April 2023; thus, most 2-RN skin assessments were conducted with the assistance of tele-ICU RNs. The academic hospital ICU had the largest increase in 2-RN skin assessment compliance, reaching 90% or higher after implementation.

The number of PIs identified as present on admission totaled 1153 (Figure 3), with a mean of 20.6 per month before implementation and 64.1 per month after implementation, equating to a 211% increase overall. The number of unit-acquired PIs totaled 324, with a mean of 15 per month before implementation and 10.5 per month after implementation, equating to a 30% decrease overall (Figure 3).

Segmented regression analysis of 2-RN skin assessment compliance (Figure 4) demonstrated a significant preimplementation monthly 1.41% increase in compliance (P < .001). Additionally, compliance increased 19.4% the first month (P < .001), which was sustained at a rate that remained higher than that of the preimplemetation period; a 0.62% increase (P = .001) occurred monthly. The number of PIs identified as present on admission increased significantly after implementation (P = .02; Figure 5). The unit-acquired PI rate per 1000 patient days decreased monthly by 0.02 after implementation, a change that was nonsignificant (Figure 6). With an interrupted time series approach, segmented regression analysis addresses variations or trends (eg, seasonal variations) that would have happened without the intervention. The results demonstrated these variations; however, during the postimplementation period, more variation occurred and unit-acquired PI rates were lower for multiple months, as compared with the preimplementation period.

The TIC DOWN-PI project optimized tele-ICU and technological advances, including early identification of present-on-admission PIs and initiation of prevention or treatment interventions. We found no previous studies that examined the impact of collaboration between tele-ICU RNs and bedside RNs to implement PI prevention best practices. Studies examining the use of telemedicine for PI prevention and management in home or non–acute care settings demonstrated weak evidence of the benefit of various technologies; however, the quality of data varied widely.^14,15  In critical care, mortality is the primary focus for measuring tele-ICU impact.¹⁶  More recent studies have explored improvements in best-practice compliance¹⁷  and standardization of care such as glycemic control, lung-protective ventilation, and venous thromboembolism prophylaxis.¹⁸  An article on tele-ICU nursing published in 2010 included examples of practice standardization such as prevention and treatment of PIs.¹⁹ 

We used an interrupted time series approach to address unit variations (eg, staffing changes, competing priorities, or other projects), differences in patient populations, and seasonal variations. Historically, at the academic medical center, PI rates have been higher during winter months. For an interrupted time series analysis, a minimum of 8 time points before and after implementation are recommended to obtain sufficient power for estimating regression coefficients.^20–22  For this project in the segmented regression, we analyzed data obtained for 10 months before implementation and 18 months after implementation (including 2 winter seasons).

Tele-ICU RNs provided an innovative approach using high-definition video technology cameras to help correctly identify PIs within 24 hours of patient admission, reducing the documentation burden for bedside RNs and the time spent finding another RN to perform a skin assessment at the bedside. By increasing 2-RN skin assessment compliance, improving documentation, and allowing for early implementation of prevention measures, PI identification upon admission reduced the designation of unit-acquired PIs and the development of future PIs. Although our project did not achieve the primary aim of a 50% reduction of unit-acquired PIs, a 30% decrease is a significant clinical improvement. Early success can be attributed to participation from key stakeholders, designated RN champions, WOCNs, and bedside leaders to promote change management, accountability, sustainability, and celebrations. Although only the academic medical center ICU met the aim of 90% compliance with 2-RN skin assessments within the first 24 hours of admission, the increased compliance of 80% was critical to the decrease in unit-acquired PIs. Achieving a compliance rate of 90% or greater may further decrease unit-acquired PIs. Due to the success of the TIC DOWN-PI project, the practice of using tele-ICU for a 4-eyes skin assessment¹⁰  has expanded to most ICUs across the health care system, and the organization’s virtual RN programs have started using the process for non-ICU departments.

Early consultation with clinical informatics experts at project initiation was integral for extracting accurate data, merging multiple data sources, and developing and distributing reports. An obstacle that we encountered was capturing compliance with PI best-practice prevention measures that were initiated within the first 24 hours of admission due to the complexity of data extraction within the EMR, REDCap survey results, and the volume of prevention interventions.

Another limitation was that our results were from only 3 ICUs in 1 health care system and therefore may not be generalizable. The segmented regression analysis used aggregate data for monthly rates rather than individualized patient data to conduct the more complex analysis. This approach led to a 1-month difference in time periods and data reporting between the unadjusted and segmented regression analyses; however, the observed difference was negligible. We did not seek input from tele-ICU or bedside ICU RNs on the impact of this practice change or the time and cost implications. Future studies could include data regarding staff perceptions of adopting technology into practice and the potential use of a WOCN in tele-ICU.

In the future, tele-ICU could provide opportunities for efficiency, collaboration, and innovation to develop best practices, improve quality of care, and provide guardrails that sustain practices across a large academic health care system. Collaboration among bedside RNs, tele-ICU RNs, and WOCNs improved compliance with 2-RN skin assessments within the first 24 hours of admission and earlier initiation of prevention best practices, resulting in a decrease in unit-acquired PIs. Results from this project will help provide evidence for the expansion of tele-ICU roles to meet patient care needs, enhancing patient safety and use of scarce resources.

The authors thank the tele-intensive care unit and bedside nurses for leading the way in this project. We appreciate their openness to collaboration outside of their departments to deliver excellent care to the communities we serve. We would like to acknowledge the champions and leaders who led the effort to improve patient safety, changed best-practice standards, and supported this project: Charlotte Gibson, MSN, RN, Lisa Konzen, MA, BSN, CCRN, Lisa Kidin, PhD, MHA, MSN, RN, CPHQ, NEA-BC, Emily Dodd, MSN, CCRN, Sharla Emery, MSN, RN, John Bell, BSN, RN, and April Hodel, DNP, RN, CCRN.