About HQI’s Sentinel Signal Detection System Report
HQI’s sentinel signal detection system automatically detects abnormal changes (i.e., signals) in the incidence of terms detected in Collaborative Healthcare and Patient Safety Organization (CHPSO) patient safety reports. Every quarter, HQI aggregates these signals across reporting hospitals to provide a statewide perspective.
Q4 2024 CHPSOData Patient Safety Signals
Q4 2024 CHPSOData Patient Safety Signals
Unhealthy Work Environments Delaying Bed Flow & Increasing Infection Risks
- Staff conflicts with Environmental Services (EVS) are slowing room turnover, causing patient bottlenecks, discharge delays, and biohazard risks.
- Actionable Insight: Gaps in teamwork may stem from differences in training, policies, and reporting structures between contract and employed staff.
Violent Patients & Weak Zero-Tolerance Enforcement
- Repeatedly violent patients continue to harm staff.
- Actionable Insight: Hospitals need clearer zero-tolerance enforcement, including protocols for removal and law enforcement escalation.
Patients Leaving AMA Without Decision-Making Capacity
- Patients eloping despite being medically unfit to make decisions.
- Actionable Insight: Hospitals need clear protocols for handling patients with impaired decision-making.
Drug Diversion Found Through Documentation Discrepancies
- Unaccounted medications flagged through inconsistent charting.
- Actionable Insight: Hospitals must strengthen documentation audits by integrating their EHR systems, dispensing cabinets, and tracking pharmacy supply.
C. Diff & Isolation Protocol Failures
- Missing health records preventing initiation of isolation protocols.
- Actionable Insight: Staff need better training on infection control documentation.
Graveyard Shift Training Gaps on High-Alert Labs
- Nurses unsure of responsibilities for anticoagulation therapies.
- Actionable Insight: Night shift staff need clearer protocols and access to immediate escalation pathways.
The Hospital Quality Improvement Platform (HQIP) provides hospital leaders with the ability to review the latest signals for their organization each quarter. In this report, these signals are aggregated across reporting hospitals and provided as counts representing a statewide signal.
As an example, consider these results for the term “intimidate”:
The Term indicates the word or word grouping reported. Natural Language Processing techniques are implemented to capture as much context as possible. For example, the words “intimidate,” “intimidates,” and “intimidated” would all be counted as “intimidate.” Pairs of words that appear will always be shown in alphabetical order.
The Positive column gives the number of reporting hospitals that had a statistically significant increase in the count of uses of this term for this quarter (i.e., the count of positive signals).
The Negative column gives the number of reporting hospitals that had a statistically significant decrease in the count of uses of this term for this quarter (i.e., the count of negative signals).
The Neutral column gives the number of reporting hospitals that did not see a statistically significant change in the count of uses of this term this quarter.
The Net Signal Value provided is the absolute difference between the number of Positive and Negative signals. The absolute difference is the difference between the two with a positive sign; a Positive Signal count of 20 and a Negative Signal count of 12 would lead to a Net Signal Value of 8 (as would a Positive Signal count of 12 and a Negative Signal count of 20).
The Most Correlated Term gives the term that most commonly occurs with the signaling term in the patient safety records.
The # category provides a bar chart with the quarterly counts of uses of this term among reporting hospitals. Hovering over each bar with a mouse cursor will provide a tooltip with the quarterly counts. Some reporting hospitals are excluded from this report if the most recent quarter’s report volume is much higher or much lower than typical for those hospitals.