Identification of Local Problem
We identified the issue of delirium assessment compliance during our 2023 Quality Verification (QVP) visit at Henry Ford Hospital. There were marked variances in the location for flowsheet documentation of delirium assessments (confusion assessment measure or CAM/CAM-ICU), which led to inconsistent interventions and challenges with auditing compliance.
Context of the QI Activity
The goal was standardization of the approach when assessing and auditing delirium scores. To understand the current state, Confusion Assessment Method (CAM) and CAM-ICU flowsheets were audited to track delirium assessment completion within our SICU population. This initial review (completed in partnership with Nursing Education), showed a 54% completion rate of the assessment along with multiple data source options and variances in clinical practice by the surgical team
What were the interventions?
These findings were shared with the Surgical Quality/Safety Committee and a subsequent review of the order-set showed that not all post-operative patients were receiving the assessment because the CAM scoring order was linked with the ventilator bundle order. A work group was built including P4 unit educator, Nurse Manager, NSQIP clinical reviewer, and Surgeon Champion with the pilot location of P4 Surgical Intensive Care unit. The work group reviewed current state for documentation, which included identifying how the CAM flowsheet populated for surgical ICU patients and if there were any deviations for inclusion. It was quickly identified that there was an issue with admission order sets that prevented patients who were not intubated from being included in the filter. The team worked with EPIC to standardize the assessment process, by modifying the placement of the CAM flowsheet to ensure it is included in all SICU admission order-sets for post-operative patients instead of only being included for those who were intubated. It was also quickly identified that delirium assessment alone would not address the problems caused by patients experiencing symptoms; although an important first step.
The workgroup then reached out to EPIC to see if there was a way to update the admission flowsheet bundle to include the Confusion Assessment M (CAM) flowsheet. The updated filter was added to the EPIC spring release April 2025. Weekly audits continued and staff education was updated in the meantime to increase compliance.
Education was also rolled out to Nursing, Residents, APPs and Physicians to review expectations and clarify inclusion populations. The workflow was released during the Spring EPIC update (April 2024).
Costs and Funding Sources
This project was not “funded,” but support was received from the Hospital Associate Chief Quality Officer, Quality Director, the SQSC group, the SICU medical director, and Nursing.
Overall Results and Analysis
The first 2 weeks of launching the flowsheet regardless of intubation status, there was a noted improvement from baseline completion. Initial audits of the CAM flowsheet averaged around 52% completion. On February 5, 2025 the CAM flowsheet order was officially added to the SICU admission order set and audits that week improved to a rate of 80% and continued to climb over the next couple weeks into the 90% range. Based on the improvements noted in the Spring update, the project was expanded to include sleep hygiene practices in addition to delirium reduction and build in the referral for early intervention.
Staff feedback was positive that the workflow was more streamlined and did not require additional adjustments to the workflow process. This ensured that documentation was completed in a timely manner and improved overall compliance.
Limitations
The main limitation that we experienced was in relation to time and available resources. Given the joint merger between Henry Ford Health System and Ascension Health system in Michigan, there were longer lead times needed to adjust workflow and the potential for projects being denied as the IT team was being prioritized to the merge.
Lessons Learned
Lessons learned for the project included the QVP process really being a driving force for change. With the need to address the delirium assessments and tracking clearly outlined in our action report, it was very clear to leadership why the project should take place. Other lessons included timing of updates, which allowed us to target education and buy in from staff. We also identified that our current process for follow up if they did test positive for delirium needed an update as well and that would continue in phase 2.
