HospitalInspections.org

Based on staff interview and record review, the hospital failed to follow the request of the legal representative for 1 of 1 patients (Patient 31) reviewed for antipsychotic medications (used to treat psychosis defined by delusions or hallucinations) to not receive antipsychotic medications.

This failure resulted in Patient 31's rights not being respected with regard to a course of treatment.

Findings:

Patient 31 was admitted on 8/12/17 with a diagnosis of dementia with behavioral disturbance.

A review of the clinical record revealed a note addressed to "Medical Staff" from the Power of Attorney (POA) dated 10/24/17 and directed, " ...All anti-psychiatric [sic] medications be withdrawn effective now and permanently."

Attached was a copy of the Health Care Directive that designated the above POA as the decision maker for Patient 31.

On 1/24/18 at 2:30 p.m., Patient 31's clinical record was reviewed with the Clinical Informatics Specialist (CIS) which revealed the following:

An 11/22/17 order for quetiapine (an antipsychotic) 25 mg (milligram- a unit of measure) to be given daily at 2100 (9:00 p.m.).

Additionally, a 12/6/17 physician order for olanzapine (an antipsychotic) 5 mg was to be given at bedtime.

On 1/24/18 at 2:30 p.m., the Medication Administration Record was also reviewed during a concurrent interview with the CIS. The CIS acknowledged there was documentation Patient 31 was receiving these medications.

A review of a document titled, "ED to Hosp- Admission", indicated olanzapine was administered 14 times from 12/6/17- 12/22/17.

A review of a document titled, "ED to Hosp- Admission", indicated quetiapine was administered 28 times from 11/22/17- 12/22/17.

According to the orders, both antipsychotic medications were discontinued on 12/23/17.

According to the document titled, "Patient Rights", revised 4/16/2016, it stipulated, "You have the right to ... Make decisions regarding medical care ... in order to give informed consent or to refuse a course of treatment." It further read, "Hospital staff and practitioners who provide care in the hospital shall comply with these directives. All patients' right apply to the person who had legal responsibility to make decisions regarding medical care on your behalf."

Based on observation and staff interview and record review, the hospital failed to provide care in a safe and secure setting when:
1. Security policies were not followed to prevent the attempted abduction of an infant from the neonatal intensive care unit for 1 of 51 infants (Patient 4), and
2. Gurneys were inadequately sanitized between patients.

These failures:
1. Did not provide adequate protection from possible abductions for vulnerable infants, and
2. Placed patients at risk for the development of healthcare-associated infections.

Findings:

1. Review of the policy titled Infant Abductions Prevention of and Response dated 1/28/16 stipulated " ...Any patient with a Child Protective Services (CPS) Hold or requested for No Information due to social issues should be considered high risk and be placed in rooms that provide the best setting for observation of visitors and patients ... All infants will be under direct supervision and/or observation through monitoring (split screen, the ability to display one patient's vital signs from their monitor onto another monitor)) by nursing personnel at all times."

During a tour of the Neonatal Intensive Care Unit (NICU) on 6/28/17 at 2:45 p.m., with Registered Nurse 2 (RN 2), Pod (a separate grouping of four neonatal bedspaces) "B" was observed to have four infant spaces. Upon entering the pod from the sliding door, the back right pod had two infants (twins) in one bed space, the other spaces had one infant each in the designated space. Pod "B" is located at the end of the "U" shaped hall, not in line of sight from the nurses' station.

During an interview with Registered Nurse 11 (RN 11), on 6/28/17 at 3:15 p.m., Patient 4 had been in the right front space of the pod and had just been discharged to foster care. RN 11 stated Patient 4 had been placed on a CPS Hold. She further stated it had been reported during change of shift report the mother of Patient 4 had attempted to take the infant from the unit around 2:30 a.m. this morning.

RN 2 later confirmed the aborted abduction during an interview on 6/28/17 at 3:45 p.m. She further stated the attempted abduction occurred around 2:45 a.m. this morning ... the mother of Patient 4 had been discharged the morning of 6/27/17 and was informed at that time the infant had been placed on a CPS hold. The mother came in early this morning to see Patient 4. Patient 4 was in a car seat undergoing a "Car seat challenge" (an assessment to determine if an infant able to sit in their car seat before discharge without any health complications). RN 2 continued to state Patient 4's mother removed the transport stroller from the back nurses' station and loaded the infant and car seat into the stroller ... she then silenced the alarms on the monitor and disconnected the infant from the monitor ... she strolled through the center, staff only, hallway and exited the unit through unlocked doors and appeared to be going towards the staff elevators. RN 2 stated the night shift charge nurse entered the hallway from the staff break room and recognized the mother and infant and redirected them back to the unit. Security was called and the mother of Patient 4 was escorted out of the building.

During a concurrent observation and interview, on 6/28/17 at 4:00 p.m. with RN 2, the back hallway was observed and all doors were tested. No card or code is required to open the doors. The door to from the NICU to the back hallway was not alarmed. The doorway to the staff elevators and the stair well emitted a low digital alarm that is not audible to the NICU. RN 2 stated there is no additional infant security system where the infant is tagged with a device that alarms when the infant is removed from the unit. RN 2 concurred had the charge nurse not exited the hallway, the mother and Patient 4 would have be able to leave the floor.


29821


2. During a 1:44 p.m., 1/23/18 observation in the Emergency Department (ED) with the department's Manager (EDM) and Administrative Nursing Director (AND), treatment room 4's gurney mattress was turned over, revealing two supply wrappers.

In a concurrent interview, the AND identified that the treatment room was patient-ready. Upon observing the wrappers, she stated, "It looks like we need to do some cleaning."

During a 1:47 p.m., 1/23/18 observation in treatment room 3, the area beneath the gurney mattress was also inspected and debris observed. The AND stated the gurney did not meet the department's standard for cleanliness.

In a 2:12 p.m., 1/23/18, RN 1 affirmed that she had just completed cleaning treatment room 29 after a patient's recent departure. She stated she usually wipes down only the top and sides of gurney mattress pads when preparing treatment rooms for incoming patients.

During a concurrent observation of the room with the EDM and AND, a brown stain was observed on the frame of the gurney below the mattress. The stain was able to be removed by a cleaning wipe.

In a 9:58 a.m., 1/25/18 observation with the EDM and AND, the area beneath the gurney mattress in treatment room 25 was inspected. Two fingernail trimmings were noted.

In a concurrent interview, the AND confirmed the debris appeared to be nail trimmings.

In a 10:12 a.m., 1/24/18 interview, the ED's Clinical Educator indicated staff hadn't been specifically trained to sanitize the undersurfaces of gurney mattresses when preparing rooms for future patients.

Review of the hospital's 1/16 "Beds, Gurneys and Wheelchairs" policy was silent on the expected procedure for sanitizing gurneys.

Based on interview, medical record and document review, the hospital failed to ensure an effective Quality Assurance and Performance Improvement (QAPI) program that investigated a serious adverse event, death of Patient 3 associated with potential opiate over administration related to an intravenous (IV) infusion pump (medical device used to deliver controlled amounts of medication through plastic catheter tubing inserted into a patient's blood vessels) failure, with depth and rigor, in accordance with manufacturer recommendations, hospital policy, and governance expectations.

A. Hospital investigators of the incident failed to:

1. Gain understanding/expertise of infusion pump continuous quality improvement (CQI) reports and event logs and codes in order to accurately interpret and analyze potential pump malfunctions (e.g., to thoroughly evaluate possible technical sources such as sensor malfunction, audible alarm malfunction, software programming error, or design flaws). Investigators relied on manufacturer assurance that Patient 3's pump model was not designed to unexpectedly permit uncontrolled gravity-driven flow of fluid through the pump system and into the patient (free flow); so unless the pump appeared damaged, a free flow event was caused by human error. Refer to A 0286, item 1.

2. Evaluate all pumps available for use to the hospital for unexplained or undetected free flow events. Refer to A 0286, item 2.

3. Ensure corrective action steps from prior free flow events as well as current incident were effective. Failure to memorialize in policy. Failure to track and measure quality performance in pump usage. Refer to A 0286, item 3.

4. Provide clear direction to pump users that the pumps could free flow without audible alarm. Refer to A 0286, item 4.

5. Thoroughly evaluate the details and pump event logs from 4 prior similar "free flow" occurrences within their regional hospital system. Refer to A 286, item 5.

6. Maintain active surveillance for suspicion that free flow or pump failure could occur without warning to any infusion pump, even when used correctly. Quality oversight and formal presentation of the pump failure analysis stalled for 5 months and action plans were delayed. Refer to A 0286, item 6.

7. Measure opiate blood level of Patient 3 when the possible unplanned over administration of morphine was identified, or during the 5 hours that followed before he died. Refer to A 0286, item 7.

8. Report the death associated with a device failure and medication error to appropriate government agencies. Refer to A 0286, item 8.

B. The hospital's administrative officials failed to ensure that an ongoing program for quality improvement and patient safety, including the reduction of medical errors, was implemented and maintained. Refer to A 0309

These failures put all patients relying on infusion pumps for medication administration in danger of unplanned free flow overdose at any time, which could lead to need for rescue, organ damage, or death. An Immediate Jeopardy to the health and safety of patients was declared on 1/24/18 at 5:58 p.m. The hospital produced an acceptable immediate action plan to reduce the risk to patients on 1/24/18 at 7:50 p.m.

Findings:

Review of the governing body bylaws, effective 1/1/18, documented the powers and responsibilities of the Board (governing body) under Article V, Section 5-7.B. to include, "To supervise the Corporation's quality improvement activities, including all aspects of patient care evaluation ..." Article X, Section 10-1.A. indicated "the Board of Directors was responsible for the quality of professional care rendered in the medical facilities it oversees." Article IX, Section 9-6.B. defined the responsibilities of the Board Safety and Quality Committee: "(1) Conduct appropriate monitoring and oversight to ensure that the quality and safety functions of the hospitals, facilities and services are being consistently and effectively performed so as to enhance the quality of patient care and assure patient safety; (2) Monitor the implementation of a quality management program; and (3) Review and recommend to the Board quality and safety improvement activities of the Corporation."

In an interview on 1/26/18 at 12:30 p.m., the hospital's Chief Executive Officer (CEO, who was responsible for the daily operations at this one facility) explained the governance structure for reporting to the Board. The hospital with its 7 affiliate hospitals report to senior managers of the Corporation, who in turn report directly to the Board. CEO acknowledged that the Board was ultimately responsible for ensuring patient quality and safety.

Based on interview, medical record and document review, the hospital failed to ensure that the Quality Assurance and Performance Improvement (QAPI) program activities tracked errors/adverse events, analyzed causes, and implemented preventive actions that included feedback learning throughout the hospital when the QAPI program investigated a serious adverse event, death of Patient 3 associated with potential opiate over administration related to an intravenous (IV) infusion pump (medical device used to deliver controlled amounts of medication through plastic catheter tubing inserted into a patient's blood vessels) failure, without the necessary depth and rigor, in accordance with manufacturer recommendations, hospital policy, and governance expectations. Hospital investigators of the incident failed to:

1. Gain understanding/expertise of infusion pump continuous quality improvement (CQI) reports and event logs and codes in order to accurately interpret and analyze potential pump malfunctions (e.g., to thoroughly evaluate possible technical sources such as sensor malfunction, audible alarm malfunction, software programming error, or design flaws). Investigators relied on manufacturer assurance that Patient 3's pump model was not designed to unexpectedly permit uncontrolled gravity-driven flow of fluid through the pump system and into the patient (free flow); so unless the pump appeared damaged, a free flow event was caused by human error.

2. Evaluate all pumps available for use to the hospital for unexplained or undetected free flow events.

3. Ensure corrective action steps from prior free flow events as well as current incident were effective. Failure to memorialize in policy. Failure to track and measure quality performance of pump usage.

4. Provide clear direction to pump users that the pumps could free flow without audible alarm.

5. Thoroughly evaluate the details and pump event logs from 4 prior similar "free flow" occurrences within their regional hospital system.

6. Maintain active surveillance for suspicion that free flow or pump failure could occur without warning to any infusion pump, even when used correctly. Quality oversight and formal presentation of the pump failure analysis stalled for 5 months and action plans were delayed.

7. Measure opiate blood level of Patient 3 when the possible unplanned over administration of morphine was identified, or during the 5 hours that followed before he died.

8. Report the death associated with a device failure and medication error to appropriate government agencies.

These failures put all patients relying on infusion pumps for medication administration in danger of unplanned free flow overdose at any time, which could lead to need for rescue, organ damage, or death. An Immediate Jeopardy to the health and safety of patients was declared on 1/24/18 at 5:58 p.m. The hospital produced an acceptable immediate action plan to reduce the risk to patients on 1/24/18 at 7:50 p.m.

Findings:

Review of facility policy "High Alert Medications, Management of" dated 6/30/16 classified IV infusions of morphine as a high alert medication. The policy indicated "High alert medications are drugs that bear a heightened risk of causing significant patient harm when they are used in error. Although mistakes may or may not be more common with these drugs, the consequences of an error are clearly more devastating to patients."

1. Failure to understand technical reports related to pump performance and a free flow event.

In a review of medication safety data on 1/23/18 at 1:35 p.m., the Director of Pharmacy (DOP) indicated that an incident was reported that a 250 milligrams (mg) bag (volume was 250 milliliters-ml) of morphine sulfate (type of high risk narcotic opiate pain medication) infusing through a pump device into Patient 3's vein emptied more rapidly than intended on 8/28/17. The expected infusion rate was 1 mg per hour and the bag containing 250 mg was discovered empty 1.5 hours after it was started. The DOP stated that an investigation was conducted by the Quality Department and an analysis was performed. The cause of the medication error or pump malfunction was never identified. "We don't know what happened."

Review of the facility provided document on 1/24/18 beginning at 9:06 a.m. with administrative staff who included the DOP, Assistant Administrator for Quality and Safety (AAQS), Chief Nurse Executive (CNE), Assistant Administrator for Women's and Children's Center (AAWC), biomedical engineering staff (BIOM 1 and BIOM 2, who serviced and maintained infusion pump equipment) indicated that possible causes of the morphine disappearance event on 8/28/17 were considered. These included leakage out of the bag or pump tubing, stealing of the morphine from staff or persons with access, cracks or damage to the pump, incorrect seating of pump tubing through the pump module, pump module door not closed, open or broken safety clamps, failure to alarm, and programming error by user. To address these possibilities, investigators inspected the bag and tubing for holes or leaks, checked trends of narcotic discrepancies accessed by staff, interviewed staff present during the discovery, pulled the pump out of service and had it visually inspected by the hospital equipment management engineers (Biomed Department). The facility provided document did not document that the pump history would be downloaded to look for prior episodes of undetected free flow.

In an interview on 1/24/18 at 1:45 p.m., AAQS stated, "We were told by [pump manufacturer] there is no such thing as a free flow alarm, free flow alarm means that the door is open." The manufacturer insisted the pumps were safe, as long as used correctly. The facility investigation group was unaware that a free flow alarm at the time of the empty bag discovery was documented on the CQI report, and this was not discussed at the facility investigation meeting.

In an interview on 1/26/18 at 2:58 p.m., Patient 3's bedside nurse, Registered Nurse 6 (RN 6), stated that he checked the pump tubing and module placement was correct when the infusion was started at midnight. RN 6 was in and out of the room after starting the infusion, returned to the room at 1:25 a.m., at which time Patient 3's wife pointed to the morphine bag indicating it looked empty. RN 6 heard no audible alarms and the green light was lit to indicate correctly infusing, that 1.5 ml had infused since it was started, the door was not open, and indeed the 250 ml bag was empty. RN 6 stopped the pump. RN 6 checked the patient, the bed and the floor for wetness and found none. No further morphine was administered.

Contradicting the information to AAQS from the manufacturer were the bedside nurse report that after carefully seating the tubing, the module door was closed at the start of the infusion close to midnight on 8/28/17, and the CQI All Infusion Detail Report run on 8/29/17 at 10:40 a.m. The CQI report recorded the start of the infusion at 12:02 a.m. The next CQI entry was 4,857 seconds (about 80 minutes) later at 1:23 a.m. with an entry for Free Flow Alarm, immediately followed by Door Closed Alarm, Door Opened, and Door Closed entries between 1:23 a.m. and 1:26 a.m. The bedside nurse discovered the empty bag when notified by Patient 3's family member around 1:25 a.m. The discovery corresponded to the CQI Codes of Infusion Restarted, Paused, Restarted and Stopped between 1:24 and 1:26 a.m. The nurse reported that no audible or visual alarms occurred when he discovered the bag to be empty, and the green light remained lit to show normal infusing operation.

Biomed test report of Patient 3's infusion pump performed on 8/29/17 showed that free flow did occur when the module door was open, and it took 8 minutes 12 seconds to drain a 250 ml bag of fluid in the free flow mode. Biomed was not able to simulate free flow in the ordinary pump setup mode with the module door closed, found no visual damage to the pump, and concluded that the controller and delivery module functioned normally, despite a controller display message, "Replace Battery." The Biomed report made no comment or interpretation about the CQI report of free flow alarm at 1:23 a.m. on 8/28/17.

In a group interview on 1/24/18 beginning at 9:06 a.m., BIOM 2 stated that the Biomed Department was asked to inspect Patient 3's infusion pump on 8/29/17, and found no problems. BIOM 2 stated that pharmacy staff normally checked what buttons a user had pressed on the infusion pump. The DOP stated that this was not done for this pump. Concurrent review of the CQI report with BIOM 1 and BIOM 2 confirmed the 8/28/17 at 1:23 a.m. free flow entry, but BIOM 1 and BIOM 2 indicated they did not know what that entry meant. They said they would check with the manufacturer.

Review of an electronic communication between the manufacturer and BIOM 1, dated 1/24/18, indicated the pump should alarm if the safety clamp was open with the door open. However, free flow could happen with the pump door closed in rare cases. "If you believe the devices weren't working as expected, or if there was a patient event in which it is suspected that an actual free flow/over infusion occurred, (customer service) can complete an investigation with the tubing, pump module and (software programming) unit." Review of the facility provided document showed no plan to send the infusion pump to the manufacturer or request the manufacturer to open an investigation.

In an interview on 1/24/18 at 6:30 p.m., AAQS stated that the pump manufacturer told the hospital that if free flow occurred, it would happen in the first 2-3 minutes from start of an infusion.

The CQI report did not document precisely when the free flow occurred, only that a free flow code was triggered at 1:23 a.m., 1-2 minutes before RN 6 discovered the empty bag. If it takes 8 minutes to empty (per Biomed testing), the free flow occurred at least 8 minutes before 1:23 a.m., and as early as shortly after the infusion was started at 12:02 a.m.

Review of Patient 3's medical record with RN 8 on 1/26/18 at 10 a.m.., indicated that Patient 3 was 72 years old with a history of advanced lung cancer. He was admitted to the hospital intensive care unit (ICU) on 8/11/18 for breathing difficulty (respiratory failure) and pneumonia (lung infection). High level care was provided in the ICU, including 2 days of assisted breathing from a machine (ventilator), surgical procedures for left leg blood clots, and medications for a heart rhythm problem. His lung condition was not expected to improve for him to go home so consultations for hospice, palliative care (treatments to relieve pain and suffering), and "comfort care" (treatment aimed at comfort rather than cure) were documented.

On 8/27/17 at 8:54 p.m., nurse's notes documented that Patient 3 was very uncomfortable with increased work of breathing and fast breathing rate. An order for morphine infusion for pain and to help the work of breathing was issued. The morphine infusion of 1 milligram per hour (mg/hr) was started at 8:55 p.m. At 9:02 p.m., significant fall in blood oxygen level and breathing rate were documented in the vital sign summary. At 9:02 p.m., Patient 3 had a 25 second seizure. The morphine was stopped. A device to apply pressure to the lungs via facemask (BiPAP, bilevel positive airway pressure) helped improve breathing rate and blood oxygen level. The blood pressure continued to drift lower but the mean blood pressure remained above 60. Patient 3 and his spouse agreed to comfort care and an order was issued at 10:30 a.m. The morphine infusion at 1 mg/hr was restarted at 12:02 a.m.

Nurse's notes documented that at 1:25 a.m. Patient 3 appeared comfortable, no pain, but the morphine bag that started with 250 mg in 250 milliliters (ml) of fluid was completely empty with IV tubing full of air. IV pump still running with green light present, no alarms present. IV states volume infused = 1.5 ml. Morphine gtt (infusion drip) turned off. MAP (mean arterial pressure, is defined as the average pressure in a patient's arteries and is considered a better indicator of perfusion to vital organs) was 62.

Vital sign recordings documented very low blood pressures (BP) following discovery of the empty morphine bag. In the 3 days prior to the event, Patient 3's mean BP (MAP) measurements were above 60, and generally ranged from 70's to 80's. At the start of morphine infusion the MAP was 65. At 1 a.m., the MAP was 62. At 1:45 a.m., the MAP was 57, and it remained in the 50's thereafter. The last MAP recorded before death was 47 at 6 a.m. Low blood pressure is consistent with morphine effect on circulation. Counter measures to the low blood pressure condition "given the situation with morphine" included BiPAP, administration of a large volume of IV fluid, and a medication to stabilized blood pressure (dopamine), despite an order prior to the morphine discovery to provide "comfort care" only. Decision not to reverse the effects of morphine with Narcan was documented in eICU Event Note by a physician at 3:50 a.m. as it would make the patient very uncomfortable. At 5:49 nurse's notes documented 10 second seizure. At 6:02 a.m., the eICU physician documented seizures, ordered a anti-seizure medication, and noted Patient 3's wife agreed to withdraw BiPAP and dopamine. At 6:33 a.m. a nurse practitioner pronounced Patient 3 to be dead.

In an interview on 1/30/18 at 10:30 a.m., the county Coroner indicated that meetings were held with hospital administrative staff to learn about the details surrounding Patient 3's death. The Coroner contacted the hospital on 1/18/18 to share that the toxicology report for Patient 3 indicated significant morphine content, a lethal dose, in Patient 3's blood. The Coroner requested the infusion pump's CQI and event log. On 1/19/18 the Coroner contacted the hospital to ask how the CQI/event log and its codes were interpreted. "We have questions about the program codes." The hospital could not explain. A meeting was arranged with the expectation that the hospital leaders and integrated quality managers would then explain. A meeting occurred on 1/24/18 with several hospital administrative staff, including pharmacy staff, but no one knew what the codes meant or could explain how to interpret the CQI/event log and codes. The Coroner stated that the hospital investigators had visually inspected the pump, but not looked at the programming. The Coroner directed the hospital leaders to look at the programming.

2. Failure to evaluate all infusion pumps for damage and undetected free flow events.

Review of a facility document dated 8/25/16 indicated a high priority alert from the smart pump manufacturer indicating "Damaged Door components May fail to Engage Anti-Free Flow Mechanism, Potentially Leading to Gravity Flow."

In an Interview on 1/24/18 at 1:30 p.m., AAQS stated that the hospital had knowledge of other incidents of "free flow" of high risk medications to patients on the same type of pump at regional sister hospitals that resulted in harm or death. Information about these incidents was shared through a regional integrated quality program that oversees the quality activities for the entire region, and reports to the governing board.

Review of a document dated October 2016 to alert staff across the region of one of these free flow incidents noted that the unanticipated free flow of IV medications/solutions can occur with the [manufacturer name] pump and the pump WILL NOT alarm. Four free flow incidents were known to have occurred, 3 of which involved insulin, morphine, and total parenteral nutrition (all high risk medications). The alert documented that causes of free flow could include improper setup by the user and damage to the pump and its parts. An inspection of all pumps and modules was completed at that hospital, and damaged pumps were removed from service or repaired.

In interviews on 1/24/18 at 1:30 p.m. and later at 6:30 p.m., AAQS confirmed that following Patient 3's morphine incident no audits were conducted at this hospital to inspect for damage and to look at CQI reports on any of the 3500 pumps available for use in the hospital.

3. Failure to ensure corrective action steps from prior free flow events as well as current incident were effective. Failure to memorialize in policy. Failure to track and measure quality performance in pump usage.

In an Interview on 1/24/18 at 1:30 p.m., AAQS stated that the hospital had knowledge of other incidents of "free flow" of high risk medications to patients on the same type of pump at regional sister hospitals that resulted in harm or death. Information about these incidents was shared through a regional integrated quality program that oversees the quality activities for the entire region, and reports to the governing board.

Review of a document dated October 2016 to alert staff across the region of one of these incidents noted that 4 free flow incidents had occurred, 3 of which involved insulin, morphine, and total parenteral nutrition (all high risk medications). The alert documented that free flow could occur and the pump "WILL NOT alarm." Causes of free flow could include improper setup by the user and damage to the pump and its parts. All partner hospitals were asked to provide the alert plus an educational handout to all pump users. Partner hospitals were asked to document completion of an online training by the nurse users of the pump.

In an interview at 6:30 p.m., CNE stated that all the above-listed learning and training steps were completed by pump users at this hospital as directed by the regional quality leaders. CNE acknowledged that these steps did not prevent Patient 3's free flow incident on 8/28/17. As corrective actions to the potential causes of Patient 3's morphine incident at this hospital in 2017, the same educational and training actions were conducted. But this time, an additional step of return demonstration of competence (hands on demonstration) to set up and operate infusion pumps correctly was included.

Review of the hospital's infusion pump training packet that was part of the facility investigative team's corrective actions to Patient 3's morphine incident did not inform students that free flow could occur without an audible alarm, or that rarely free flow could occur with the module door closed.

Review of 2016 and 2017 training materials related to free flow incidents showed that they were very similar.

During an interview on 1/26/18 at 2:58 p.m., RN 6 who started Patient 3's morphine infusion on 8/27/17 stated that he was given training on the infusion pump in 2016 and again in 2017. RN 6 stated that he did not learn anything new in the second training.

As AAQS indicated in the 1/24/18 at 6:30 p.m. interview, no audits of CQI data to identify unidentified incidents of free flow in infusion pumps following the facility corrective action re-education and training steps were conducted to verify effectiveness of the actions. AAQS confirmed that facility policies and procedures did not address the infusion pump free flow potential. AAQS stated that no documentation was required by nursing staff to visualize the drip chamber to ensure no free flow was taking place after starting an infusion. AAQS stated that the facility was not tracking any data regarding monitoring of staff or free flow indicators on infusion pump CQI reports.

Review of facility policy "Medication Administration (Adult and Pediatric)" dated 10/1/14 documented, "Use of the ... I.V. Pump." The policy did not address any of the following: free flow potential, to check for pump damage or visualize drip chamber after starting pump (steps covered in the pump training).

Review of the pump manufacturer guidance manual, under a section titled System Reporting Module Alarms, a Free Flow Alarm should send an outbound message to the Knowledge Portal for infusion technologies, an action that could be captured and measured as a quality indicator for monitoring all pumps. The hospital presented no documented evidence that this information was used as a quality indicator for infusion pump safety.

4. Failure to provide clear communication that the infusion pump may free flow without an audible alarm.

In an interview on 1/26/18 at 2:58 p.m., Patient 3's bedside nurse, RN 6, stated that he checked the pump tubing and module placement was correct when the infusion was started at midnight, was in and out of the room after starting the infusion, returned to the room at 1:25 a.m., at which time Patient 3's wife pointed to the morphine bag indicating it looked empty. RN 6 heard no audible alarms and the green light was lit to indicate correctly infusing, that 1.5 ml had infused since it was started, the door was not open, and indeed the 250 ml bag was empty.

Review of a document dated October 2016 to alert staff across the region of unanticipated free flow of IV medications/solutions documented that free flow can occur and the pump WILL NOT alarm.

Review of the hospital's infusion pump training packet that was part of the facility corrective actions to Patient 3's 8/28/17 morphine incident did not inform students that free flow could occur without an audible and/or visual alarm.

5. Failure to evaluate the details and pump CQI reports from 4 prior similar "free flow" occurrences within their regional hospital system.

In an interview on 1/26/18 at 12:30 p.m., CEO explained that he and some hospital managers report to a higher level of regional corporate managers, including a quality division that oversees adverse events related to patient care for 8 affiliate hospitals, including this one. CEO said that his bosses (who report directly to the governing board) and higher level quality managers were aware of this free flow incident, even though it had not yet advanced beyond the local hospital quality committee for formal review by governance committees and the higher level managers. CEO stated that he expected the persons responsible for conducting the free flow investigation, with assistance from regional entities, to perform research on pump device technology, on national reports and trends to learn what others have learned, and to compare events among the hospitals to learn more. With that understanding his team should develop effective corrective actions. CEO was sure this had been done.

However, in a group interview with hospital leaders on 1/24/18 at 6:30 p.m., the Director of Risk Management (DRM, who served at the affiliate level) was asked about the 4 other similar free flow events at the sister hospitals, and what had been learned. DRM replied that she would need to look it up. There was no evidence the hospital utilized data from the prior sister hospital events to develop corrective actions to Patient 3's free flow event. The hospital leaders were also asked if the pump manufacturer who insisted the pumps were safe and would not free flow spontaneously, shared their quality control testing data with the facility quality team investigators. DRM stated, "I believe we made that request, have not received any data, don't have it."

6. Failure to maintain active surveillance for suspicion that free flow could occur without warning to any infusion pump, even when used correctly. Quality oversight and formal presentation of the investigation team's analysis stalled for 5 months and action plans were delayed.

In an interview on 1/24/18 at 9:06 a.m., CNE stated that the facility could not come up with what happened in Patient 3's care. CNE stated that a physician did not believe Patient 3 could have received the full 250 mg of morphine based on the vital sign record. The County Coroner had not released their investigation results with toxicology to show whether a large amount of morphine was in Patient 3's body at the time of death.

Review of the facility provided document for Patient 3's morphine free flow incident investigation documented that the incident occurred on 8/28/17 and the investigation was complete on 9/5/17. As discussed above in items 1-5, several areas were not understood or thoroughly investigated before closing the investigation. Yet, investigators stated that they were not sure if the pump failed, if free flow occurred without user error, or if Patient 3 received significant amount of morphine to contribute to Patient 3's death.

In an interview on 1/26/18 at 12:30 p.m., the hospital's CEO stated that he was informed about the investigation of Patient 3's morphine free flow incident. "We weren't sure if it was free flow, machine failure, design failure, or human error. Still not sure." CEO indicated that he and the hospital leadership judged the corrective actions to be sufficient, expected them to be effective because of the additional return demonstration step for nurse users of the pumps. However, CEO acknowledged that additional evidence surfaced this week with the CQI free flow entries and a verbal report from the County Coroner that Patient 3 did indeed receive significant amount of morphine, to increase the likelihood that unexpected free flow not due to human error may have occurred. CEO agreed that additional safety measures to prevent another free flow occurrence in other infusion pumps were indicated.

In a group interview on 1/24/18 at 6:30 p.m., AAQS stated, "We determined the pump was the problem."

Further review of the free flow facility provided analysis indicated a high-impact plan to acquire locking devices to prevent stealing of narcotic infusions. Plan to implement was first quarter 2018.

In an interview on 1/24/18 at 3:05 p.m., AAQS stated that the hospital was thinking of using a device to limit the amount of fluid being infused (volutrol) to minimize the amount of medication reaching a patient should free flow occur. But that plan was not implemented due to a shortage of the product.

In an interview on 1/25/18 at 9:30 a.m., CNE and AAQS indicated that the facility quality team's analysis was scheduled to go to Council Care Standards, Quality Committee and Medical Executive Committee, but had not yet moved up for higher level reviews. The high-impact action had not yet been implemented.

Review of policy titled "Patient Safety Reports: High Risk Event and Unusual Occurrences Reporting and Management," dated 8/8/17, documented, "All activities of the facility quality team analysis will be presented by members of the team to the Quality Council and Patient Safety Committee. Relevant recommendations and actions will be reviewed by appropriate committees (Medical Staff Committees, Council on Patient Care Standards, etc.) ...Findings and recommendations from the analysis or significant/never event analysis will be reported to the appropriate Medical Staff Committee, Quality and Patient Safety Committee, Medical Executive Committee, and the Medical Policy Committee."

7. Failure to measure Patient 3's opiate level after potential overdose of morphine was identified and Patient 3 exhibited low blood pressure requiring rescue interventions.

Review of Patient 3's medical record with RN 8 on 1/26/18 at 10 a.m., indicated that RN 6 documented very low blood pressures (BP) following discovery of the empty morphine bag on 8/28/17. In the 3 days prior to the event, Patient 3's mean BP (MAP) measurements were above 60, and generally ranged from 70s to 80s. At the start of morphine infusion the MAP was 65. At 1 a.m. the MAP was 62. At 1:45 a.m. the MAP was 57, and it remained in the 50's thereafter. The last MAP recorded before death was 47 at 6 a.m. Low blood pressure is consistent with morphine effect on circulation. Counter measures to these effects included high pressure oxygen flow through a mask device attached to the face (Bilevel Positive Airway Pressure, BiPAP), administration of a large volume of IV fluid, and a medication to stabilized blood pressure, despite an order prior to the morphine discovery to provide "comfort care" only. On 8/28/17 at 6:33 a.m. a nurse practitioner pronounced Patient 3 dead. The clinical record indicated there were no orders, lab results, or blood draws to measure morphine toxicity levels following the identification of the empty morphine bag, or in the following 5 hours prior to his death. Potential exposure to a lethal dose of morphine 5 hours prior to death was not documented in Patient 3's death summary.

In a group interview on 1/24/18 at 10:15 a.m., CNE stated that two physicians said it was impossible for the patient to have received 250 mg of morphine and survive for another 5 hours. The hospital had the ICU doctor and an independent physician look at the case and both said there was no way the patient got 250 mg morphine. The County Coroner was notified of possible medication error related to device malfunction. The Coroner investigation has not yet been released.

In an interview on 1/30/18 at 10:30 a.m., the County Coroner reported that toxicology results confirmed that a significant, lethal dose, amount of morphine was measured in Patient 3's body. The Coroner said that they were contacted by a nurse because the hospital was unsure if a pump malfunction or free flow administration of 250 mg of morphine infused over a short time, and that could be related to Patient 3's death. The Coroner investigation was not yet completed.

The hospital had doubts whether Patient 3 was affected by the overdose of morphine when the Coroner referral was made, as well as earlier when the event occurred. Not measuring Patient 3's blood for morphine toxicity was a missed opportunity to determine how much morphine was received, and whether it played a role in Patient 3's subsequent course and death.

8. Failure to report suspected pump device failure to government agencies.

In an interview on 1/24/18 at 2:45 p.m., AAQS stated that the Risk Department chose not to report Patient 3's death associated with a medical device because a physician felt there wasn't any way that Patient 3's death was caused by morphine. If the Coroner toxicology report pointed toward the morphine, then Risk would classify it differently.

Review of hospital policy titled "Patient Safety Reports: High Risk Event and Unusual Occurrences Reporting and Management," dated 8/8/17, documented "California Department of Public Health Reportable Events ... Medication error-patient death/disability."

Review of hospital policy titled "Performance Improvement (PI) and Patient Safety (PS) Program 2017", last approved 5/10/17, documented on page 5: "Any reportable adverse event under Health and Safety Code 1279.1 including ... 6. Patient death or serious disability associated with the use or function of a device in patient care in which the device is used or functions other than as intended. For purposes of this subparagraph, "device" includes, but is not limited to, a catheter, drain, or other specialized tube, infusion pump, or ventilator .... 11. A patient death or serious disability associated with a medication error, including, but not limited to, an error involving the wrong drug, the wrong dose, the wrong patient, the wrong time, the wrong rate, the wrong preparation, or the wrong route of administration, excluding reasonable differences in clinical judgment on drug selection and dose."

In an interview on 1/24/18 at 2:45 p.m., AAQS indicated that the corporate staff was responsible to report incidents of device failure to the United States Food and Drug Administration (FDA).

Review of the FDA online requirements for reporting potentially unsafe medical devices, retrieved at https://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm#requirements, with webpage heading Medical Device Reporting, indicated that mandatory reporters (i.e., manufacturers, device user facilities, and importers) were required to submit certain types of reports for adverse events and product problems to the FDA about medical devices. In addition, the FDA also encouraged health care professionals, patients, caregivers and consumers to submit voluntary reports about serious adverse events that may be associated with a medical device, as well as use errors, product quality issues, and therapeutic failures. These reports, along with data from other sources, can provide critical information that helps improve patient safety. Under Device User Facilities: A "device user facility" is a hospital, ambulatory surgical facility, nursing home, outpatient diagnostic facility, or outpatient treatment facility, which is not a physician's office. User facilities must report a suspected medical device-related death to both the FDA and the manufacturer.

Based on interview and record review the hospital's administrative officials failed to ensure that an ongoing program for quality improvement and patient safety, including the reduction of medical errors, was implemented and maintained when analysis of an adverse event, the over administration of an opiate medication to Patient 3 via infusion pump (a device intended to control the rate of fluid flowing from a reservoir bag through plastic catheter tubing into a patient's vein), overlooked critical corrective actions and corrective actions were delayed by not advancing investigation findings to the oversight committees in a timely manner. This failure had the potential to result in any patient in the hospital to receive an over administration of any medication delivered via infusion, which could cause harm up to and including death.

Findings:

Review of Patient 3's medical record with Registered Nurse 8 (RN 8), on 1/26/18 at 10 a.m., indicated that Patient 3 was 72 years old with a history of advanced lung cancer. He was admitted to the hospital intensive care unit ICU) on 8/11/18 for breathing difficulty (respiratory failure) and pneumonia (lung infection). High level care was provided in the ICU, including 2 days of assisted breathing from a machine (ventilator), surgical procedures for left leg blood clots, and medications for a heart rhythm problem. His lung condition was not expected to improve for him to go home so consultations for hospice, palliative care (treatments to relieve pain and suffering), and "comfort care" (treatment aimed at comfort rather than cure) were documented.

On 8/27/17 at 8:54 p.m., nurse's notes documented that Patient 3 was very uncomfortable with increased work of breathing and fast breathing rate. An order for morphine infusion for pain and to help the work of breathing was issued. The morphine infusion of 1 milligram per hour (mg/hr) was started at 8:55 p.m. At 9:02 p.m., significant fall in blood oxygen level and breathing rate were documented in the vital sign summary. At 9:02 p.m., Patient 3 had a 25 second seizure. The morphine was stopped. A device to apply pressure to the lungs via facemask (BiPAP, bilevel positive airway pressure) helped improve breathing rate and blood oxygen level. The blood pressure continued to drift lower but the mean blood pressure remained mean arterial pressure, (MAP, is defined as the average pressure in a patient's arteries and is considered a better indicator of perfusion to vital organs) above 60. Patient 3 and his spouse agreed to comfort care and an order was issued at 10:30 a.m. The morphine infusion at 1 mg/hr was restarted at 12:02 a.m.

Nurse's notes documented that at 1:25 a.m., Patient 3 appeared comfortable, no pain, but the morphine bag that started with 250 mg in 250 milliliters (ml) of fluid was completely empty with IV tubing full of air. IV pump still running with green light present, no alarms present. IV states volume infused = 1.5 ml. Morphine gtt (infusion drip) turned off. MAP was 62.

Vital sign recordings documented very low blood pressures (BP) following discovery of the empty morphine bag. In the 3 days prior to the event, Patient 3's MAP measurements were above 60, and generally ranged from 70's to 80's. At the start of morphine infusion the MAP was 65. At 1 a.m., the MAP was 62. At 1:45 a.m., the MAP was 57, and it remained in the 50's thereafter. The last MAP recorded before death was 47 at 6 a.m. Low blood pressure is consistent with morphine effect on circulation. Counter measures to the low blood pressure condition "given the situation with morphine" included BiPAP, administration of a large volume of IV fluid, and a medication to stabilized blood pressure (dopamine), despite an order prior to the morphine discovery to provide "comfort care" only. Decision not to reverse the effects of morphine with Narcan was documented in eICU Event Note by a physician at 3:50 a.m. as it would make the patient very uncomfortable. At 5:49, nurse's notes documented 10 second seizure. At 6:02 a.m., the eICU physician documented seizures, ordered an anti-seizure medication, and noted Patient 3's wife agreed to withdraw BiPAP and dopamine. At 6:33 a.m., a nurse practitioner pronounced Patient 3 to be dead.

Review of Patient 3's medical record with RN 8, on 1/26/18 at 10 a.m., indicated that RN 6 documented very low blood pressures (BP) following discovery of the empty morphine bag on 8/28/17. In the 3 days prior to the event, Patient 3's mean BP (MAP) measurements were above 60, and generally ranged from 70's to 80's. At the start of morphine infusion the MAP was 65. At 1 a.m. the MAP was 62. At 1:45 a.m. the MAP was 57, and it remained in the 50's thereafter. The last MAP recorded before death was 47 at 6 a.m. Low blood pressure is consistent with morphine effect on circulation. Counter measures to these effects included high pressure oxygen flow through a mask device attached to the face (Bilevel Positive Airway Pressure, BiPAP), administration of a large volume of IV fluid, and a medication to stabilized blood pressure, despite an order prior to the morphine discovery to provide "comfort care" only. On 8/28/17 at 6:33 a.m. a nurse practitioner pronounced Patient 3 dead.

In an interview on 1/30/18 at 10:30 a.m., the County Coroner stated that a nurse from the hospital contacted the County Coroner on 8/28/17 with concern that the infusion pump may have failed when a 250 mg bag of morphine emptied in a short period (1.5 hours), 5 hours before Patient 3 died. The Coroner's toxicology report confirmed that Patient 3 received a lethal dose of morphine.

In a group interview on 1/24/18 beginning at 9:06 a.m., the Administrative Assistant for Quality and Safety (AAQS) and Chief Nurse Executive (CNE) indicated that a facility quality analysis of the unexpected rapid free flow of morphine into Patient 3 without the pump audibly alarming led to corrective actions to try to prevent another occurrence of free flow. AAQS indicated that the facility quality team did not notice that a continuous quality improvement (CQI) infusion device report of the pump software program documented a free flow alarm at 1:23 a.m. close to the time RN 6 discovered the empty morphine bag. Although the infusion pump was visually inspected by the hospital's equipment engineers, neither the pharmacist, the engineers, nor the nurse and quality leaders understood how to interpret the CQI entry for free flow alarm. The facility quality team did not determine the cause of the free flow but chose education and training as corrective actions for potential human error as the cause.

In a subsequent interview on 1/24/18 at 6:30 p.m., AAQS indicated that the facility quality team did not send the pump to the manufacturer and request an investigation from them for potential device flaws, did not inspect other infusion pumps for evidence of unnoticed free flow alarm entries on software reports, did not obtain details and data from sister hospitals where 4 similar free flow events had been shared with hospital leaders in October 2016, and most importantly, did not implement additional safeguards to limit the volume of medication that could infuse if a pump should fail again.

Review of hospital policy titled "Performance Improvement (PI) and Patient Safety (PS) Program 2017", last approved 5/10/17, documented that the PI Plan focused on indicators related to improved health outcomes and the prevention and reduction of medical errors to ensure that patient care, processes and outcomes were measured, analyzed, and improved. The PI Plan described the structure for oversight of patient safety from measurement and analysis of data from multidisciplinary committees, adverse event analysis and action plans, etc. reported to Quality/Safety committees and Medical Staff committees, and ultimately to the Governing Board.

In an interview on 1/24/18 at 9:06 a.m., CNE stated that the facility could not come up with what happened in Patient 3's care. CNE stated that a physician did not believe Patient 3 could have received the full 250 mg of morphine based on the vital sign record. The County Coroner had not released their investigation results with toxicology to show whether a large amount of morphine was in Patient 3's body at the time of death.

In an interview on 1/26/18 at 12:30 p.m., the hospital's Chief Executive Officer (CEO) stated that he was informed about the investigation of Patient 3's morphine free flow incident. "We weren't sure if it was free flow, machine failure, design failure, or human error. Still not sure." CEO indicated that he and the hospital leadership judged the corrective actions to be sufficient, expected them to be effective because of the additional return demonstration step for nurse users of the pumps. However, CEO acknowledged that additional evidence surfaced this week with the CQI free flow entries and a verbal report from the county Coroner that Patient 3 did indeed receive significant amount of morphine, to increase the likelihood that unexpected free flow not due to human error may have occurred. CEO agreed that additional safety measures to prevent another free flow occurrence in other infusion pumps were indicated.

In a group interview on 1/24/18 at 6:30 p.m., AAQS stated, "We determined the pump was the problem."

Review of the facility provided document for Patient 3's morphine free flow incident investigation documented that the incident occurred on 8/28/17 and the investigation was complete on 9/5/17. The investigation was closed without a clear understanding of how to analyze the infusion pump software programming codes, alarm codes, or determine the cause of the free flow incident that affected Patient 3. In multiple interviews, hospital managers stated that they were not sure if the pump failed, if free flow occurred without user error, or if Patient 3 received significant amount of morphine to contribute to Patient 3's death.

Further review of the free flow facility provided document indicated a high-impact plan to acquire locking devices to prevent stealing from narcotic infusions. Plan to implement was first quarter 2018.

In an interview on 1/24/18 at 3:05 p.m., AAQS stated that the hospital was thinking of using a device to limit the amount of fluid being infused (volutrol) to minimize the amount of medication reaching a patient should free flow occur. But that plan was not implemented due to a shortage of the product.

In an interview on 1/25/18 at 9:30 a.m., CNE and AAQS indicated that the facility quality analysis was scheduled to go to Council Care Standards, Quality Committee and Medical Executive Committee, but had not yet moved up for higher level reviews. The high-impact action had not yet been implemented.

Review of policy titled "Patient Safety Reports: High Risk Event and Unusual Occurrences Reporting and Management," dated 8/8/17, documented, "All activities of the RCA will be presented by members of the team to the Quality Council and Patient Safety Committee. Relevant recommendations and actions will be reviewed by appropriate committees (Medical Staff Committees, Council on Patient Care Standards, etc.) ...Findings and recommendations from the root cause analysis or significant/never event analysis will be reported to the appropriate Medical Staff Committee, Quality and Patient Safety Committee, Medical Executive Committee, and the Medical Policy Committee."

Further review of the 2017 PI Plan documented "Objective F: To assure an effective communication system for reporting performance improvement activities throughout the region, including to the Boards, Medical Staffs, hospital staff, and Administrations." Objective H directed to "...create a culture of safety" by creating processes that encouraged "Organizational learning about safety errors including knowledge sharing to effect behavioral changes within this and other health care organizations."

In an interview on 1/26/18 at 12:30 p.m., CEO explained that he and some hospital managers report to a higher level of regional corporate managers, including a quality division that oversees adverse events related to patient care for 8 affiliate hospitals, including this one. CEO said that his bosses (who report directly to the governing board) and higher level quality managers were aware of this free flow incident, even though it had not yet advanced beyond the local hospital quality committee for formal review by governance committees and the higher level managers. CEO stated that he expected the persons responsible for conducting the free flow investigation, with assistance from regional entities, to perform research on pump device technology, on national reports and trends to learn what others have learned, and to compare events among the hospitals to learn more. With that understanding his team should develop effective corrective actions. CEO was sure this had been done.

However, in a group interview with hospital leaders on 1/24/18 at 6:30 p.m., the Director of Risk Management (DRM, who served at the affiliate level) was asked about the 4 other similar free flow events at the sister hospitals, and what had been learned. DRM replied that she would need to look it up. There was no evidence the hospital utilized data from the prior sister hospital events to develop corrective actions to Patient 3's free flow event. The hospital leaders were also asked if the pump manufacturer who insisted the pumps were safe and would not free flow spontaneously, shared their quality control testing data with the facility quality team investigators. DRM stated, "I believe we made that request, have not received any data, don't have it."

Based on interview and document review, the medical staff failed to ensure an effective system for determining the clinical privileges granted to practitioners for sedation care, critical care, and emergency care were in accordance with accepted professional standards for high quality care as directed by the medical staff bylaws when:

1. Privileging requirements for moderate and deep sedation (the administration of sedative and narcotic medications to lower consciousness and help patients tolerate painful procedures) "carved out" exceptions from maintaining formal emergency rescue training and certification (e.g. advanced life support (ALS) standardized steps to rescue patients from states of heart and lung function cessation) to 107 of 107 critical care, cardiovascular, and emergency medicine practitioners (Medical Doctors-MDs 3-109) without verifying current knowledge, experience, and competence for rescue procedures relevant to the patient population they routinely cared for;

2. Privileging requirements for 8 of 8 pediatric (MDs 3-10) and 10 of 10 neonatal (newborn) (MDs 11-20) critical care practitioners departed from nationally recognized organizational guidance for verifying emergency rescue training and certification for those practitioners when clinical privileges were requested;

3. Privileging requirements for ultrasound-guided procedures for 8 of 8 pediatric critical care (MDs 3-10) and 33 of 33 emergency medicine (MDs 21-53) providers lacked specificity for current experience and competence measures based on published guidance from nationally recognized organizations.

These failures put patients at risk for adverse outcomes from care and procedures performed by unqualified practitioners.

Findings:

Review of the 3/2/17 Medical Staff Bylaws documented under Article II, item 2.01-1(a) that the Medical Staff's purposes shall be to assure that "all patients receive care at a uniform level of quality and efficiency consistent with generally accepted standards." Item 2.01-2(a) noted that the Medical Staff's rights and responsibilities shall include "to provide quality patient care," and "to account to the Board of Directors for the quality of patient care by all Members authorized to practice in the hospital." Under Article III, item 3.04(a) required each Member of the Medical Staff to provide patients with care of the generally recognized professional level of quality and efficiency. Article VI detailed the basis for privilege determinations under item 6.04-2, "The Medical Staff shall make an objective and evidence-based decision with regard to each request for Privileges. Requests for Privileges shall be evaluated on the basis of the Practitioner's education, training, experience, demonstrated professional competence, ..." Item 6.03 documented, "each Department will be responsible for developing criteria for granting Clinical Privileges. The criteria shall be designed to facilitate uniform quality patient care, treatment and services."

1. Sedation privileging departed from national standards.

Review of the Procedural Sedation Privilege List for Non-Anesthesiologists form with the Medical Staff Director (MSD) on 2/24/18 at 2 p.m. documented under Moderate Sedation Criteria item (b), "Current ACLS or PALS certification, including airway management*" as a requirement at appointment to the medical staff. ACLS referred to emergency rescue training for adults and PALS referred to emergency rescue training for children. A requirement at reappointment to the medical staff for this privilege noted, "Current ACLS or PALS certification, including airway management course.*" The asterisk explanation further documented exceptions for emergency medicine and critical care practitioners; the exempt providers were not required to show current emergency rescue training, nor to perform the required 10 moderate sedation cases every 2 years to show current experience of using the skill. Cardiovascular surgeons who performed 10 intubations (insertion of a breathing tube through the mouth into the large airway) in prior two years were exempt from the emergency rescue training requirement. Cardiologists were exempt from emergency rescue training. Deep Sedation (deeper loss of consciousness approaching a state of general anesthesia) privileges were restricted to emergency medicine and critical care practitioners who satisfied the Moderate Sedation requirements (though exempted as described) and documented 10 successful administrations of Deep Sedation in prior two years.

Review of a hospital-wide policy titled "Procedural Sedation (Moderate and Deep) Administration, last approved 7/1/15, noted that the hospital intended to provide "uniform, safe, and evidence-based standards for the management of all patients who are receiving sedation for the purpose of completion of a procedure." Under the Physician Provider Requirements section, the physicians must be privileged. No requirement for current emergency rescue training and certification was listed. However, under the Nurse Competency Requirements section, current emergency rescue certification was required of nurses. No current standards of practice references or other evidence-based sources were documented on the policy.

Review of the American Society of Anesthesiologists (ASA) Statement on Granting Privileges for Administration of Moderate Sedation to Practitioners Who Are Not Anesthesia Professionals, last amended 10/19/11, documented under item 11 for Education and Training that the non-anesthesiologist sedation practitioner granted moderate sedation privileges will have age-relevant advanced life support skills and current certificate from a vendor that included "hands-on" training and skills demonstration of airway management and automated external defibrillator (AED) use. The same requirement was documented under Education and Training item 3.13 in the ASA Advisory on Granting Privileges for Deep Sedation to Non-Anesthesiologist Sedation Practitioners, approved 10/20/10.

In an interview on 1/26/18 at 10:30 a.m., the Medical Doctor (MD) 2 (who served as Chief of Anesthesia and was responsible for ensuring safe sedation care and related sedation policies throughout the hospital) stated that he expected anesthesia and sedation practices, privileging requirements, and policies to be based on guidance from the ASA. MD 2 did not know of any particular national organizational standards that provided "carve-out" exceptions for emergency rescue training and certification for emergency medicine and critical care practitioners, as the ASA had not done so in their guidance statements. MD 2 stated that his Anesthesia Board Certification provided periodic practice and competency testing for airway management and rescue skills. Yet all members of the Anesthesia Department also kept current on ALS certification apart from Board testing.

Review of the medical staff file for MD 3 with MSD on 1/24/18 at 2 pm indicated that MD 3 was granted moderate and deep sedation privileges at reappointment on 5/5/16. The file did not show documented evidence of a current ALS certificate, or verification of the number of moderate and deep sedation cases performed in prior two years. Airway management proficiency was documented by current maintenance of Board certification in Pediatric Critical Care Medicine, but documented evidence of current ALS training was lacking in the file.

MSD provided a list of medical staff profiles related to sedation and selected procedure privileges on 1/25/18. Of the 8 Pediatric Intensive Care Unit (PICU) physicians, 2 had evidence of PALS certification (MD 6 and 9), MD 5 had Board certification that expired 2/15/16, and MD 8 did not have Board certification. All 8 PICU providers were granted sedation privileges or other privileges to perform procedures that required moderate or deep sedation (such as placement of plastic catheters into deep arteries and veins, or drainage of fluid from the chest or abdomen).

In an interview on 1/25/18 at 2 p.m., MD 4 (PICU physician) stated he was uncertain whether emergency rescue certification was required for sedation privileges, and deferred to the Anesthesia Department to set the requirement.

Further review of the list of medical staff profiles related to sedation and selected procedure privileges showed that medical staff files for the 10 Neonatal Intensive Care Unit (NICU) physicians captured only 2 emergency rescue certificates (MD 12 and 20). However, in an interview on 1/25/18 at 2 p.m., MD 11 stated that all (10 of 10) NICU physicians did possess current Neonatal ALS certificates, as it was required by their private medical group and MD 11 was an administrator of the group.

In an interview on 1/26/18 at 10:30 a.m., MD 2 (Anesthesia Chief) indicated he was not familiar with the ALS requirements for the private emergency medicine group or whether all of them were certified for adult and pediatric emergency rescue, given they all attend to both age groups in their routine work. MD 2 was not familiar with the curriculum content for Board certification of the critical care and emergency medicine Board organizations, as specifically related to emergency rescue procedures, i.e. whether they were comparable to vendors such as the American Heart Association as advised by the ASA. MD 2 acknowledged that some critical care and emergency medicine physicians may not maintain current Board certification in their specialties, but may still have been granted sedation privileges.

Additional review of the medical staff profile list also showed that of the 33 emergency medicine providers, the medical staff files captured only 4 (MD 21, 28, 33 and 35) with current adult emergency rescue certificates, and none with current child rescue certification. Of the 28 adult critical care providers granted sedation privileges, medical staff files captured 9 (MD 54, 55, 57, 61, 66, 76, 77, 78, and 81) with current adult emergency rescue certificates. Of the 28 cardiovascular providers with sedation privileges, the medical staff files captured 4 (MD 82, 85, 93 and 94) with current adult emergency rescue certificates.

Review of the Code Blue Log for 2017 documented 4 PICU patient codes attended by 2 PICU physicians (MD 4 x 1, MD 6 x 2), 1 NICU physician (MD 12), and 3 emergency medicine physicians (MD 25, MD 44, and MD 50) for the entire year. Only MD 6 had documented certification for pediatric ALS on file with the medical staff. Of the 155 adult patient codes, the vast majority were attended by ED, cardiovascular, and pulmonary physicians lacking documented evidence of ALS certification.

In an interview on 1/26/18 at 2:30 p.m., MD 1 (who served as the Chief of Staff to oversee all medical staff members) stated that hospital physicians were expected to provide high quality patient care based on guidance from nationally recognized organizations as well as published literature, which served as the basis for professional standards of practice. MD 1 acknowledged that the hospital-wide Procedural Sedation (Moderate and Deep) Administration policy did depart from published standards (ASA and others) by not requiring physicians to be formally trained and certified in emergency rescue care. MD 1 indicated that the "carve-out" exceptions were decided locally by the medical staff, and not based on published organizational or evidence-based standards. An assumption was made that the carve-out physicians receive comparable ALS training through their Board certification programs and exercise rescue skills daily. Yet, no attempt to directly verify either of these assumptions was integrated into the credentialing process for granting and renewing sedation privileges. MD 1 acknowledged that for providers lacking formal ALS certifications, the medical staff made no attempt to verify that each member who possessed sedation privileges actually performed emergency rescue procedures to show current experience and competence when sedation privileges were renewed.

2. Privileges to care for PICU and NICU patients did not require current ALS training and certification, as recommended by organizational standards.

Review of the April 2012 American Academy of Pediatrics (AAP) guidance statement for pediatric hospital care of all levels (including critical care and neonatology care) titled "Medical Staff Appointment and Delineation of Pediatric Privileges in Hospitals," documented that the purpose for credential verification and granting of clinical privileges was to ensure that members of the medical staff were competent and qualified to provide specified levels of patient care. Credentialing Principles noted that criteria for specific clinical privileges must be well defined and based on up-to-date national and local standards. At initial appointment AAP listed documents to verify that included "Basic life support or other lifesaving (emergency rescue) course (eg, pediatric advanced life support, neonatal resuscitation program) documentation of completion. When a practitioner wishes to perform a major new procedure or treatment modality not inherent in their residency or fellowship training (such as use of ultrasonography or providing sedation), he/she must document sufficient hands-on-training to be deemed competent. Data from some new procedures have revealed that the complication rate decreases significantly and competency increases significantly after a certain number of the procedures are performed. Guidelines for competency in new procedures or treatment modalities must be developed on the basis of a review of the literature and the technical aspects of the procedure. The minimum information required from the practitioner at bi-annual reappointment included "Current cardiopulmonary resuscitation or other lifesaving course (eg, pediatric advanced life support, neonatal resuscitation program) documentation of completion. Such credentialing verifications were one way for hospitals to help ensure that their patients receive quality care."

In an interview on 1/24/18 at 2 p.m., the MSD confirmed that ALS resuscitation or lifesaving course completion was not a requirement for any pediatric providers to be granted privileges to practice in the three levels of care, general pediatrics, pediatric critical care, or neonatal critical care.

In an interview on 1/25/18 at 2 p.m., MD 4 (who served as an administrator for pediatric services) stated that current certification from an ALS lifesaving rescue training course was not a requirement for pediatric privileges. MD 4 was not certain if it was a requirement for sedation privileges. MD 4 acknowledged that pediatric practices were largely based on guidance from AAP and their specialty subgroups, but was not familiar with AAP recommendations for competency and qualifications for clinical privileges. MD 4 cited no other expert organization to guide the Pediatric Department ' s decisions for developing criteria for privileging. The Pediatric Chair was not available for interview.

At the same 1/25/18 at 2 p.m. interview, a member of the neonatology group stated that all 10 neonatologists were currently certified in resuscitation, as it was a requirement of the private medical group.

As discussed in item #1, the medical staff had no system for verifying to verify rescue certificates for pediatric department providers.

3. Privileges for central venous catheter insertion (placing a plastic catheter tube into a major vein) and drainage of chest and abdomen fluid did not distinguish whether they were to be performed with ultrasound (US) guidance. In addition it did not specify or how the qualifications and competence for US guided would differ from non-US guided procedures, as recommended by organizational standards.

Review of the privilege requirements for emergency medicine providers on 1/24/18 at 2 p.m. with MSD indicated that Point of Care Ultrasound for any purpose (resuscitative, diagnostic, symptom/sign-based, procedure guided, therapeutic/monitoring) had separate and distinct qualifications from emergency core privileges. Point of Care Ultrasound (also known as bedside ultrasound) required formal training through a residency/fellowship program or course attendance or possession of the privilege at another hospital. Oversight for the first six cases was required (proctoring). However, no requirement to perform a minimum number of ultrasound experiences was necessary to continue the privilege every two years at reappointment. No requirement to meet ongoing competency reviews for image quality, accuracy of clinical interpretation, or comparison to actual patient outcome was documented as part of the determination for whether a provider was competent to continue bedside ultrasound care without oversight.

The hospital presented no evidence of specific data-driven review of bedside ultrasound competency by emergency medicine providers that was integrated into the decision to grant bedside ultrasound privileges at reappointment.

Review of the American College of Emergency Physicians (ACEP) Policy Statement Emergency Ultrasound Guidelines, approve October 2008, documented detailed expectations for bedside ultrasound training, qualifications, experience, and competency for physicians in training and for established emergency physicians who had not received formal training. ACEP emphasized the growing need and variety of uses for bedside ultrasound. For example, diagnosis can be hastened, testing can be more selective, and for certain procedures it can improve patient safety. "Hospitals focusing on patient safety have also embraced national and international safety guidelines that strongly recommend the use of ultrasound in central venous access; a critical procedure with significant potential complications that was routinely performed 'blindly' before the clinical use of ultrasound." ACEP provided the following guidance:
· Hands-on training must be incorporated for Core Emergency (bedside) Ultrasound applications (included Procedural Guidance).
· Methods of determining competency include traditional testing, testing using simulator models, videotape review, observation of bedside skills, over-reading of images by experienced sonologists (expert physicians who perform and interpret ultrasound examinations), and monitoring of error rates through a quality assurance process.
· Performance improvement programs that monitor accuracy will help to ensure that quality ultrasound studies are being performed.
· Hospitals and their medical staffs are legally obligated to credential and re-credential those physicians who can demonstrate current clinical competence, skill, judgment, and technique. In order to remain competent, physicians must stay current with the literature and perform services delineated in their clinical privileging on a regular basis.
· The objective of the QA (Quality Assurance) process is used to evaluate images for technical competence and interpretations for clinical accuracy and provide feedback to improve physician performance.
· The QA process is an integrated part of the educational, training, and credentialing processes of each department.
· Parameters to be evaluated might include image resolution, anatomic definition, and other image quality acquisition aspects such as gain, depth, orientation, and focus. In addition, the QA system should compare the impression from the emergency ultrasound interpretation to the patient outcome measures such as consultative ultrasound, other imaging modalities, surgical procedures, or patient clinical outcome.
· All images obtained prior to a physician becoming credentialed should be reviewed, while images from credentialed physicians may be sampled during QA.
· An important step to managing risk is insuring that physicians are properly trained and credentialed according to national guidelines such as those set by ACEP.

ACEP inferred that a minimum number of bedside ultrasound examinations be performed to determine ongoing competence: Each health care provider must perform and interpret minimal ultrasound examinations per credentialing cycle as designated by the ACEP Ultrasound guidelines. Health care providers who do not meet this number requirement must have their studies over-read by a qualified health care provider (such as US Director).

In an interview on 1/26/18 at 2:30 p.m., MD 1 (Chief of Staff) stated that ultrasound guidance for central venous catheter placement was now the standard. MD 1 expected that all providers with privileges to place central lines to be utilizing ultrasound. Such providers would include emergency medicine, pulmonology, cardiovascular, and critical care. MD 1 acknowledged that the privilege requirements for central venous catheter placement and other procedures that utilize ultrasound (e.g. drainage of chest and abdomen) did not require verification of training in bedside ultrasound for non-recent grads or minimum number to perform each cycle. MD 1 acknowledged that quality reviews for reappointment did not routinely evaluate performance (in a manner similar to ACEP QA tasks) or verify the frequency (volume) of performing bedside ultrasound procedures, QA did not verify image quality, accuracy of clinical interpretation, or comparison to actual patient outcome with data to integrate into decisions to renew bedside ultrasound procedures at reappointment. The primary indicator used to drive renewal of privileges was reported adverse events related to a specific procedure. If none were reported, the practitioner was assumed to be competent.

Review of privileging requirements for central line placement and drainage of the chest and abdomen for pediatric critical care providers with MSD, on 1/24/18 at 2 p.m., indicated no language that bedside ultrasound was to be used for those procedures. Therefore, no education/training qualifications were incorporated into those privileges. No proctoring, oversight, quality reviews specific to bedside ultrasound technique, or volume to maintain competence were defined as necessary to show competence for renewal of those procedure privileges. The medical staff office made no attempt to confirm training in bedside US for critical care providers. MSD confirmed that the number of procedures was not verified at reappointment, and that no attempt to distinguish procedures that did or did not utilize bedside ultrasound was made.

In an interview on 1/25/18 at 2 p.m., MD 4 (who had pediatric critical care privileges including central venous catheter placement, drainage of chest and abdomen) stated that his unit acquired a bedside ultrasound device about a year ago and he now routinely used it for procedures, as did other critical care providers in his unit.

Review of the medical staff file for MD 3, a pediatric critical care provider, on 1/24/18 at 10:30 a.m. with MSD, indicated that MD 3 had privileges renewed on 5/31/16 for the following:

Drainage of chest - required 2 cases at reappointment, attested to 5
Placement of central venous catheters - required 1 case at reappointment, attested to more than 30
Placement of catheter into artery - required 1 case at reappointment, attested to 10
Drainage of abdomen - required 1 case at reappointment, attested to 2

MSD presented a case volume report that listed the number of hospital admissions and selected indicators, but the list did not specify the number of patients who had these 4 procedures done by MD 3. The case volume report did not sort out the use of bedside US. MSD stated that no attempt to drill down on privilege-specific activity was made at the time of reappointment to determine experience and competence for those privileges. Data from Department selected indicators did not capture volume or performance for those specific privileges for the pediatric department, and neither for other critical care departments.

Review of the current Pediatric Department Rules and Regulations indicated no rule to guide privilege requirements and performance evaluations to be based on current standards of practice or organizational standards for education, training, experience, and competence.

Review of the American College of Surgeons (ACS) Revised Statement on Recommendations for Use of Real-Time Ultrasound Guidance for Placement of Central Venous Catheters, published online February 1, 2011, documented that the use of US has been associated with a reduction in complication rate and an improved first-pass success when placing catheters in the neck and upper body veins. The technique also decreased procedural time and complication rate. Standardization of education, training, and practice was an important component of the technique. ACS referred to the 2001 Agency for Healthcare Research and Quality recommendation for US guided placement of central venous catheters to be one of the top 11 evidence-based practices that health care providers can use to improve patient care and patient safety. ACS also referred to guidance from the National Institute for Clinical Excellence, which included appropriate training to achieve competence for providers who performed this task.

Review of the medical staff roster indicated 33 emergency medicine providers, 28 adult critical care providers, 28 cardiovascular providers, 8 pediatric critical care providers, and 10 neonatal critical care providers. In an interview on 1/26/18 at 4 p.m., the Assistant Director of Quality indicated that US guided procedures were not performed in neonatal care, but were performed routinely by physicians in these other departments. Therefore, the medical staff failed to have a system to verify experience and competence on 97 providers who were granted US guided procedural privileges.

Based on staff interview, medical record and document review, the hospital failed to safely provide nursing services when:
A. Drugs were not administered to two of 56 Sampled Patients according to physician orders, resulting in a morphine overmedication via an intravenous (IV, into a vein) infusion pump of 166 times the prescribed dose (Patient 3) and exposure to heightened risk of bleeding associated with the blood thinner heparin (Patient 50) (Refer to A 405), and
B. A policy and procedure for caring for adult patients placed on the Extracorporeal Membrane Oxygenation device (ECMO, a medical device that pumps and oxygenates blood outside the body, allowing the heart and lungs to rest) outside the Cardiovascular Surgery Intensive Care Unit (CVSICU) was not developed for two Sampled Patients (Patients 43 and 56) (Refer to A 386).

An Immediate Jeopardy (IJ) situation was declared at 5:58 p.m., 1/24/18 for unresolved and potentially ongoing failure in infusion pumps hospital-wide, potentially affecting all patients using the devices. Facility leadership team members including the Chief Executive Officer (CEO), Chief Medical Executive (CME), Chief Nurse Executive (CNE), Regional Executive, Surgery Services (AESS), Assistant Administrator of Integrated Quality Services (AAQS) and Assistant Administrator for Women and Children's Center (AAWC) were notified of the IJ at 6:30 p.m., 1/24/18. The leadership team were notified that the IJ was abated at 7:50 p.m., 1/24/18 after an immediate corrective action plan was accepted by the Department.

These failures placed patients at risk for inconsistent and substandard care, injury and harm, including death.

The cumulative effect of these systemic failures resulted in the inability of the hospital to comply with the Condition of Participation: Nursing Services.

Based on staff interview, medical record and document review, the hospital failed to:
1) Develop policy and procedure for providing care to adult patients on the Extracorporeal Membrane Oxygenation device (ECMO; a medical device that pumps and oxygenates blood outside the body, allowing the heart and lungs to rest) outside the Cardiovascular Surgery Intensive Care Unit (CVSICU) for two of 56 Sampled Patients (Patients 43 and 56);
2) Reassess pain with vital signs and before and after the administration of pain medication per policy for three of 56 Sampled Patients (Patients 24, 26 and 27); and
3) Ensure safe administration of medications when the effect of medication administered for Patient 6 was not monitored in accordance with hospital policy.

These failures placed patients at risk for inconsistent and substandard care when delineation of care and nursing responsibilities for patients with ECMO in place were not outlined in a policy and procedure, had the potential to cause patients to unnecessarily experience untreated pain, and put patients at increased risk of undue suffering when the effect of intended therapy was not monitored and assured.

Findings:

1) During an interview with the Chief Nursing Executive on 1/26/18 at 10:30 a.m., the CNE confirmed Patients 43 and 56 were on the ECMO device while in the Cardiac Intensive Care Unit (CICU) and in the Medical Intensive Care Unit (MICU), respectively.

Clinical record review on 1/25/18 indicated Patient 43 was in the CICU with ECMO in place from 10/17/17 through 11/13/17. Registered Nurses (RNs) who took care of Patient 43 were trained CICU nurses but were not trained in caring for a patient on the ECMOdevice.

In an interview with Registered Nurse 9 (RN 9) on 1/26/18, at 2:05 p.m., she confirmed she took care of Patient 43 while she was in the CICU. RN 9 stated as an ICU nurse, she was "not trained to take care of patients on ECMO; but 2 Pediatric Intensive Care Unit (PICU) ECMO RNs were responsible for managing Patient 43's ECMO machine".

In an interview with RN 10 on 1/26/18, at 2:15 p.m., she confirmed she took care of Patient 43 while she was on ECMO in the CICU. RN 10 said, "I'm not trained to take care of patients on ECMO machine."

During an interview with the Administrative Director of Hospital Operations (ADHO) on 1/26/18, at 4:45 p.m., she acknowledged the hospital does not have a policy and procedure for caring for adult patients on ECMO device outside of the CVSICU. The ADHO said, "CICU/MICU nurses who took care of ECMO patients did not get a formalized training [in caring for ECMO patients]."

The hospital's CVSICU policy and procedure titled "Extra Corporeal Membrane Oxygenation (ECMO)" dated 7/1/13, indicated, "...The ECMO device itself will be attended at all times by a trained perfusionist [a medical professional trained and educated to setup and operate a heart-lung machine]. The perfusionist will be responsible for the adjustment and maintenance of the pump...". The policy further indicated bedside nusre responsibilities prior to the patient's arrival. It further stipulated
the following assessments:"...Initial Assessment "follow normal routine Cardiac Surgery Post-Op Management", as well as guidelines for Ongoing Assessment, Interventions, Reportable Conditions and Documentation. It further included an Addendum on the procedure for Sterile ECMO Cannula Dressing Changes.

The hospital's policy and procedure titled "NICU/PICU: ECMO (Extracorporeal Membrane Oxygenation) Infants and Children, Management of" dated 2/1/16, indicated, "Who May Perform: Registered RN who has completed ECMO Bedside Nursing Class or ECMO Clinical Specialist Training...". The policy further delineated responsibilities for the bedside nurse for the care assessment of the patient, and the responsibilities for the ECMO specialist for the care of the ECMO circuit/device.

2A) Review of Patient 24's electronic medical record (EMR) revealed she presented in the Emergency Department (ED) with worsening abdominal pain at 9:49 p.m., 1/22/18. In triage at 10:19, 1/22/18, she described her pain as an "8" at rest and a "9" with activity on a 0 - 10 scale; on this scale, "10" is the worst pain imaginable. An Emergency Severity Index (ESI, 5-level patient categorization scale based on illness severity and resource needs, a "1" is most urgent) of "3" was assigned.

In the initial hours after Patient 24's arrival, no pain relief medication was documented as having been given. She was noted to have continuing abdominal pain at 12:51 a.m., 1/23/18, but no quantification of pain intensity was documented. At 1:05 a.m., 1/23/18, a physician order was given to medicate Patient 24 with morphine for pain; she received the medication at 1:16 a.m., 1/23/18.

The EMR did not reflect a reassessment of Patient 24's discomfort level after the morphine was received, but a physician order was given for another pain medication, hydromorphone, at 2:44 a.m., 1/23/18. She was taken for a radiological procedure and received the drug after at 3:35 a.m., 1/23/18 for a pain score of "8" at rest.

At 3:58 a.m., 1/23/18, Patient 24 was noted to have ongoing abdominal pain of an undocumented intensity. Vital signs (blood pressure, heart and respiratory rates, and temperature measurements) were taken at 4 a.m., 6 a.m., 7 a.m. and 11 a.m., 1/23/18 but pain assessments were not documented.

Patient 24 was taken to the interventional radiology suite (area within x-ray department in which invasive procedures are performed) at 11:08 a.m., 1/23/18 for a biopsy (surgical procedure to obtain a tissue sample) which was performed at 12:16 p.m., 1/23/18. Prior to the beginning of the procedure, a pain evaluation using staff objective criteria was initiated at 12:28 p.m., 1/23/18. The patient was assessed to be pain-free.

Thus, a pain assessment was not quantified and/or documented between 10:19 p.m., 1/22/18 - 3:35 a.m., 1/23/18 and between 3:35 a.m., 1/23/18 - 12:28 p.m., 1/23/18.

In an 11:49 a.m., 1/24/18 concurrent interview and EMR review, the ED Administrative Nursing Director (AND) stated that hospital policy required comfort assessment prior to the administration of pain medication and within one hour after its receipt.

2B) Patient 26's EMR indicated she arrived in the ED at 7:49 a.m., 1/24/18. At 7:55 a.m., 1/24/18, vital signs were taken and the patient self-assessed her pain level to be a "0". Patient 1 was assigned an ESI of "3".

A 1:30 p.m., 1/24/18 an EMR entry read, "Patient c/o [complaining of] body aches and chills." The intensity of Patient 26's pain was not quantified. Vital signs were retaken at 3 p.m., 1/24/18 but did not include a pain reassessment.

At 5:26 p.m., 1/24/18, Patient 26 was given Norco (pain medicaion) for discomfort. No pain reassessment had been documented since 7:55 a.m., 1/24/18. Reassessment for the effectiveness of the Norco was not done until 7:21 p.m., 1/24/18 when Patient 1 rated her pain as a "2" at rest and a "6" with activity.

In a 2:35 p.m., 1/25/18 interview, the AND noted that as a result of Patient 26's ESI of 3, her pain should have been reassessed with vital signs "at least every 4 hours." In addition, the AND stated Patient 1's pain should have been reassessed prior to giving pain medication and within an hour after its administration.

2C) Patient 27 arrived in the ED at 2:58 a.m., 1/24/18. She was pain-free in triage at 3 a.m., 1/24/18. Her assigned ESI level was "2."

Vital signs were retaken at 4 a.m., 5 a.m., 6 a.m., 7 a.m., 7:15 a.m. and 8 a.m. without documented pain reassessments; her discomfort was reevaluated at 8:11 a.m., 1/24/18.

Review of the hospital's 9/27/16 "Pain Management" policy revealed, "Pain should be routinely monitored, assessed, reassessed, and documented clearly to facilitate treatment and communication among health care clinicians...Evaluate pain and document intensity rating with routine vital signs...Evaluate and document the effectiveness of pain relief after one-time and prn (not routine, scheduled) medications along with any untoward effects within one hour following each pain-relieving intervention...."


16276


3) On 1/24/18 at 10:30 a.m., the medical record for Patient 6 was reviewed with Registered Nurse (RN 5). One of Patient 6's symptoms being treated was pain. Patient 6 has a physician order to administer ibuprofen (a pain medication that also reduces fever) every 6 hours as needed: The order dated 1/21/2018, was to administer ibuprofen 100 milligrams every 6 hours as needed for fever higher than 38.3 degrees or for pain score 1-5 (10 point scale). A 10-point pain assessment scale is a method for assessing pain intensity with zero indicating no pain and 10 indicating worst possible pain.

Patient 6 was 18-month-old baby. Concurrent review of Patient 6's record showed administration of ibuprofen, the following three doses were selected for evaluation:

1/22/18 at 1726 (5:26 p.m.)
1/22/18 at 2246 (10:46 p.m.) and on
1/23/18 at 4:46 a.m.

RN 5 specified that the dose administered on 1/22/18 at 5:26 p.m., was for elevated temperature, not for pain.

Concurrent evaluation of the record showed no documented temperature or pain assessment before or after administering ibuprofen, which was ordered to be given when needed for pain or elevated temperature, for the three doses selected.

No documented temperature before or after ibuprofen administration on 1/22/18 at 5:26 p.m.

No pain assessment before ibuprofen administration on 1/22/18 at 10:46 p.m.

No pain assessment before or after ibuprofen administration on 1/23/18 at 4:46 a.m.

The facility's policy titled, "Pain Management (adult and Pediatrics)" revised on 12/2016, was reviewed. Under procedure for pain assessment, the policy stated the following: "If pain is present, a more comprehensive pain assessment should be done to include pain intensify and location." The policy listed different pain scales including (FLACC) the Faces, Legs, Activity, Cry, Consolability observation tool that is used for patients age 2 month to 7 years who cannot self-report pain. The policy also indicated to "evaluate and document the effectiveness of pain relief after one-time and prn medications along with an untoward effects within one hour following each pain relieving intervention and on as needed basis by the RN (Registered Nurse) ...."

Patient 6's record had no documented evidence of pain assessment before or after interventions (administering ibuprofen) for 5 out of six possible opportunities associated with the three doses reviewed.

Based on staff interviews, document and record reviews, the facility failed to ensure that medications were administered as ordered by the physician when:

1) Patient 3 was administered 250 milligram (mg) of IV (intravenously, injected through the vein) morphine (a potent narcotic for pain) over one and a half hours instead of the prescribed 1 mg per hour via a smart infusion pump (medical device used to deliver medications into a patient's body in a controlled manner instead of free flow where drug can be delivered in uncontrolled large quantities in a short time. Smart pumps are equipped with safety features, such as user-alerts that activate when the user sets the pump's parameters outside of specified safety limits). This failure exposed Patient 3 to effects of morphine overdose (166 times the prescribed dose), including low blood pressure and death.

An immediate jeopardy (IJ) situation was declared on 1/24/18 at 5:58 p.m. pertaining to unresolved and potentially ongoing failure in the infusion pump, with a potential of affecting all patients using the device. The facility leadership including Chief Executive Officer (CEO), Assistant Administrator of Integrity Quality Services (AAQS), Chief Nurse Executive (CNE), Regional Executive Surgery Services (AESS), Assistant Administrator for Women's and Children's Center (AAWC) and Chief Medical Executive (CME) were verbally notified of the Immediate Jeopardy situation at 6:30 p.m. on 1/24/18. The leadership team were notified that the IJ was abated at 7:50 p.m. on 1/24/18 after an immediate corrective action plan was accepted by the Department. The corrective action plan included the following measures: 2-3 minute visual check of all infustions not using a volutrol by two registered nurses when tubing is changed or pump door is opened; every one hour check of volume and documentation on patient flow sheet; staff education to begin immediately.

2) Patient 50 was not administered heparin (blood thinner to prevent blood clots) per Physician orders. This failure had the potential of exposing the patient to side effects of the medication including bleeding.

Findings:

1) Review of Patient 3's clinical record indicated that he was admitted to the facility on 8/11/17 for acute respiratory failure (inability of the lungs to maintain normal respiratory function). Patient 3 needed high flow oxygen into the nose to ensure oxygen delivery to blood and organs.

Review of Patient 3's physician orders, dated 8/27/17 at 3:45 p.m., indicated "morphine 250 mg in NaCl 0.9% (saline solution) 250 ml (milliliter) IV Drip (premix) CONTINUOUS (continuously administered at the specified rate) ...initial dose 1 mg/hour [1ml/hr is equal to 1 mg of morphine/hr]."

Review of Patient 3's MAR (medication administration record) indicated morphine drip was started on 8/27/17 at 8:54 p.m. and stopped at 9:02 p.m. (8 minutes later).

Review of Patient 3's nursing notes dated 8/27/17 indicated:
- At 9:02 p.m.: "pt [patient] with decrease RR [respiratory rate], oxygen saturation down to 70's. Morphine gtt [drip, infusion] turned off."
- At 9:07 p.m.: "Dr ...returned page. Updated on situation. New order received to start on BIPAP (bi-level positive airway pressure is a type of device that helps with breathing) ..."
- At 9:27 p.m. "Dr ...at bedside. Discussing situation with wife."
- At 9:50 p.m. "Decision made by wife with Dr ...present to transition to comfort care."

Review of Patient 3's physician orders 8/27/17 at 10:52 p.m. indicated "morphine 250 mg in NaCl 0.9% (saline solution) 250 ml (milliliter) IV Drip (premix) CONTINUOUS."

Review of Patient 3's MAR for 8/27/17 indicated morphine drip was restarted at 11:59 p.m. with a comment "Pt now on comfort care, wife ready for morphine ...to start." Patient 3's morphine drip was stopped on 8/28/17 at 1:25 a.m. (1 hour 26 minutes later) and the comment indicated "gtt stopped pump without any alarms bag empty."
Review of nurses' note on 8/28/17 at 1:25 a.m. indicated "Upon checking IV pump, noticed morphine gtt was completely empty with IV tubing full of air. IV pump still running with green light present, no alarms present. IV states volume infused =1.5 ml. Morphine gtt tuned off, charge RN(Registered Nurse) ...on unit and notified ...(charge RN) checked pump, correctly set with tubing fully inserted into channel correctly ...BP (blood pressure)= 85/52 (normal blood pressure for a healthy adult is 120/80)."

Review of nurse's note on 8/28/17 at 2:30 a.m. indicated current blood pressure 75/50.

Review of FDA information for morphine under warnings indicated "Hypotensive (low blood pressure) Effect ... may cause severe hypotension in an individual whose ability to maintain their blood pressure has been compromised by depleted blood volume ...Morphine must be injected slowly; rapid intravenous administration may result in chest wall rigidity ..." Morphine carries a black box warning which the strongest is warning the US FDA (United States Food and Drug Administration) can require a manufacturer to put on the drug.

Review of Patient 3's vital signs (blood pressure, heart rate etc.) indicated that his MAP (mean or average arterial blood pressure during a single cardiac cycle. MAP of at least 60 is necessary to perfuse the coronary arteries that supply blood to the heart, brain, and kidneys. Normal range is around 70 - 110) was above 60 before the morphine infusion was started on 8/27/17. Review of Patient 3's MAP after the morphine drip was stopped on 8/28/17 at 1:25 a.m., indicated that it had dropped below 60. Review of Patient 3's physician orders, dated 8/28/17 at 4:15 a.m., indicated IV Dopamine (used to treat low blood pressure) 0-10 microgram/kilogram/minute to be titrated every 10 minutes for a goal of MAP between 60 and 65.

Review of physician notes dated 8/28/17 at 3:50 a.m. Indicated "RN (registered nurse) called to report 250 mg bag of morphine possibly given to the patient over one hour ...May use BIPAP to support breathing." Review of nurses notes dated 8/28/17 at 5:55 a.m. indicated "New orders received to stop pressors (dopamine) and remove BIPAP." Nurse's note dated 8/28/17 at 6:12 a.m. indicated "BIPAP removed."

On 8/28/17 at 6:33 a.m. Patient 3 was pronounced dead. The clinical record indicated there were no orders, lab results, or blood draws to measure morphine toxicity levels following the identification of the empty morphine bag. Review of Patient 3's discharge summary (summary of the major treatments and events while hospitalized) dated 8/28/17 at 6:58 a.m. did not address morphine administration.

The above findings were acknowledged by AHSO (Administrative Director of Hospital Operations) during interviews on 1/24/18 at 9:20 a.m. and on 1/26/18 at 1:51 p.m. and on 1/26/18 at 10 a.m. with RN 8.

During an interview on 1/26/18 at 2:58 p.m. RN 6 stated that on 8/27/17 he started Patient 3's morphine infusion (11:59 p.m. per MAR) in the smart pump by checking that the tubing was in the correct place, cassette was in chamber correctly, the infusion was programmed correctly and verified it with another nurse. RN 6 stated that he checked Patient 3's vitals his blood pressure, respiratory rate and they "looked good". RN 6 said he was in and out of Patient 3's room. RN 6 stated that at 1:25 a.m. when he returned Patient 3's wife pointed to the morphine bag indicating it looked empty. RN 6 verified that the bag was empty but no alarms were sounding. RN 6 notified another nurse and called physician. RN 6 stated he checked the floor and the bed to see if any of the morphine solution had leaked out but was unable to find anything wet.

Review of facility policy "High Alert Medications, Management of" dated 6/30/16 classified IV infusions of morphine as a high alert medication. The policy indicated "High alert medications are drugs that bear a heightened risk of causing significant patient harm when they are used in error. Although mistakes may or may not be more common with these drugs, the consequences of an error are clearly more devastating to patients."

A group interview was held on 1/24/18 at 9:06 a.m. with Director of Pharmacy (DOP), Director of Education (DOE), AAQS, CNE, AHSO, AAWC, BIOM (Bio Med Staff) 1 and BIOM 2. CNE stated that the facility could not come up with what happened in Patient 3's case. CNE stated that the physician did not feel Patient 3 could have received the full 250 mg of morphine based on his vital signs. The infusion pump was taken to the Biomed (Biomedical engineering is the application of the principles and problem-solving techniques of engineering to biology and medicine) department who ran the detailed report and no problem was identified in the pump. The DOP stated they looked for any diversion indicators but were unable to find any. CNE stated that there was no coroner's report received at present. AAQS stated that the event was not classified as a sentinel (unanticipated event in a healthcare setting resulting in death or serious physical or psychological injury to a patient not related to the natural course of the patient's illness) event but a facility quality analysis was done.

In the group interview BIOM 2 stated that they received the pump and using the manufacturer software they tested it and did a visual inspection and were not able to replicate the error. BIOM 2 stated that pharmacy normally checks what buttons a user pressed on the infusion pump. DOP stated that this was not done for this pump.
Review of the Biomed produced detailed report (CQI report) of the smart pump activity indicated an entry on 8/28/17 at 12:02 a.m. indicating infusion started, and on 8/28/17 at 1:23 a.m. indicating Infusion alarmed, free flow (when infusion can be injected into the patient at an uncontrolled rate) alarm, 4857 seconds (1 hour 21 minutes, which coincided with the time between Patient 3's morphine infusion restarting and discovering the morphine bag empty). BIOM 1 and BIOM 2 were unable to explain the "free flow alarm" part of the CQI report. They indicated they will check with manufacturer.

Review of manufacturer communication with the hospital, dated 1/24/18, indicated "Free Flow alarm" is seen in the CQI data if the pump door is open and the safety clamp is open while the infusion set is primed. Manufacturer indicated the pump should alarm if the safety clamp is open with the door open. The document indicated free flow could happen with the pump door closed in rare cases. The document indicated "If you believe the devices weren't working as expected, or if there was a patient event in which it is suspected that an actual free flow/over infusion occurred, Customer Advocacy can complete an investigation with the tubing, Pump module and PC Unit [sic - the main pump component that allows customization of infusions]." The facility biomed department had sequestered (isolated) the pump but it had not been sent to the manufacturer.

During the group interview, AAQS stated that this information about a free flow alarm for 4857 seconds was not discussed at the RCA. The group was unaware that a free flow alarm was indicated on the detailed infusion pump report. AAQS confirmed that no audits were conducted to look at detailed infusion pump reports to check for free flow issues.

In the group interview, CNE stated an alert for free flow concerns for the same manufacturer pump was received by the facility in 8/2016 via shared learning (process through which facts are shared with other hospitals in the system) and staff were provided training on the infusion pump safety based on this alert.

Review of a facility document dated 8/25/16 indicated a high priority alert from the smart pump manufacturer indicating "Damaged Door components May fail to Engage Anti-Free Flow Mechanism, Potentially Leading to Gravity Flow."

Review of a facility provided document dated 10/16 indicated "The unanticipated free flow of IV medications/solutions can occur with the ...pump and the pump WILL NOT alarm. There have been 4 events at ...affiliates in which IV medication/solutions have flowed via the ...pump. Three of these events involved high risk IV medications such as ...morphine ...Patient harm has occurred."

Review of the facility training document dated 10/2016 indicated "It has been identified that free flow of IV medications/solutions can occur in the following circumstances: Tubing not loaded properly, Tubing, or other obstruction, caught in the door-may cause small gap even if door is able to be closed . Cracked or damaged hinges, outer or inner door-may result in door not closing properly, leaving a gap ...Check for pump damage, if you see any damages or if the pump is dropped take the pump out of service ..."

Review of the facility RCA addressing Patient 3's morphine medication error dated 9/5/17 indicated " ... pump design may allow for flow of IV fluids without alarm if tubing is not properly placed or door is not properly closed. Mandatory education was done 10-11/2016 and shared learning went out to all leaders at all facilities."

Review of facility training document which was provided to staff after the 8/28/17 incident involving Patient 3 had very similar information. The training documents did not indicate that free flow could happen and the pump will not alarm.

During an interview on 1/26/18 at 2:58 p.m. RN 6 who started Patient 3's morphine infusion stated that he was given training on the infusion pump in 2016 and again in 2017. He stated that he did not learn anything new in the second training.

During the IJ group meeting on 1/24/18 at 6:30 p.m. AAQS confirmed that facility policies and procedures did not address the infusion pump free flow potential. AAQS stated that no documentation was required by nursing staff to visualize the drip chamber to ensure no free flow was taking place after starting an infusion. AAQS stated that the facility was not tracking any data regarding monitoring of staff compliance or free flow indicators on infusion pump reports. Review of the smart pump manufacturer document dated 10/17 indicated that a "Free Flow Alarm" was a reportable alarm to the manufacturer.

Review of facility policy "Medication Administration (Adult and Pediatric) dated 10/1/14 It did not address any of the above training.

Review of facility policy "Patient Safety Reports: High Risk Event and Unusual Occurrences Reporting and Management", dated 8/8/17, indicated, "All activities of the RCA will be presented by members of the team to the Quality Council and Patient Safety Committee. Relevant recommendations and actions will be reviewed by appropriate committees (Medical Staff Committees, Council on Patient Care Standards, etc.) ...Findings and recommendations from the root cause analysis or significant/never event analysis will be reported to the appropriate Medical Staff Committee, Quality and Patient Safety Committee, Medical Executive Committee, and the Medical Policy Committee."

Review of the facility RCA addressing Patient 3's morphine medication error dated 9/5/17 indicated "High Impact: Acquire locking devices for narcotic drips ...anticipated 1st quarter 2018." During an interview on 1/25/18 at 9:30 a.m. with CNE and AAQS they indicated that the RCA was scheduled to go to Council Care Standards, Quality and Medical Executive Committees but had not been done.

During an interview on 1/24/18 at 2:45 p.m. AAQS stated risk department did not report to CDPH because physician involved felt there wasn't any way that patient death was caused by morphine. AAQS stated that if toxicology review came back from Coroner then would classify it differently.

Review of facility policy "Patient Safety Reports: High Risk Event and Unusual Occurrences Reporting and Management" dated 8/8/17 indicated "California Department of Public Health Reportable Events ...Medication error-patient death/disability."

During an interview on 1/25/18 at 4:10 p.m. AAQS stated that RO (Affiliate Risk Officer) was told verbally by the coroner that Patient 3's toxicology report indicated "large amount of morphine."

In an interview on 1/26/18 at 9:45 a.m., the CNE and AAQS stated that the hospital owned 3500 infusion pumps, the same type of pump involved in the over administration (free flow) of morphine to Patient 3. They stated that at any given day 350 to 500 were being used.

In an interview on 1/26/18 at 12:30 p.m., the Chief Executive Officer (CEO) indicated that the conclusions of the facility quality analysis for Patient 3's death potentially related to a medication error did consider free flow but no one was certain that it occurred. The investigators considered free flow, machine failure, design failure, diversion, and potential human error. The hospital leaders judged that additional nurse training with return demonstration was sufficient to address the human error. The Biomed technicians and manufacturer did not confirm device failure so they assumed free flow had not occurred. They could not prove diversion. The hospital leaders were aware of 3 other similar incidents at sister hospitals over the previous few years. For some reason, it was concluded that free flow did not occur in this patient. Remedies for the other incidents were also directed at human error and retraining nurses. No further corrective actions were undertaken. Hospital leaders were aware that additional steps using a limited volume device (volutrol) could add protection from recurrence of pump free flow. However, there was a shortage of volutrol equipment for all sister hospitals to institute such a corrective action. The CEO acknowledged that during the Department survey this week, the team identified a printout of Patient 3's infusion pump that showed over 4000 seconds of free flow which had not been identified by the hospital RCA investigators. Comparisons of free flow printouts for the previous incidents were not described in the shared learning reports or the RCA for this incident at this hospital. The CEO reported that research and deep analysis of the previous incidents had been conducted by the umbrella quality management program investigators, and served as the basis for the first corrective retraining. The hospital presented no documented evidence of such a deep analysis by investigators following this incident, given knowledge that similar incidents had been reported around the country. The CEO also acknowledged that yesterday the County Coroner verbally stated that Patient 3 received a significant amount of morphine, confirming actual administration of excessive morphine prior to death. The CEO acknowledged that with the additional evidence learned this week, pump failure was more likely and additional steps (use of volutrol for high risk medications) was needed to minimize risk and recurrences.

2) Review of Patient 50's clinical record with RN 7 and RN 4 on 1/24/18 at 1:10 p.m., indicated that Patient 50 was on ECMO (Extracorporeal membrane oxygenation, a treatment that uses a pump to circulate blood through an artificial lung back into the bloodstream of a very ill baby. This system provides heart-lung bypass support outside of the baby's body).

Review of Patient 50's Heparin orders, ACT (Activated Clotting Time a test of coagulation that is used to monitor the blood thinning effect of heparin) and drip rates on 2/27/17 indicated the following discrepancies that were confirmed in concurrent interviews by RN 7 and RN 4:

-On 2/27/17 at 11:54 a.m., the Heparin rate was increased to 0.6 ml/hour instead of 0.5 ml/hour based on the ACT and physician order.
-On 2/27/17 at 12:39 p.m., the Heparin rate was reduced from 0.6 to 0.5 ml/hour instead of leaving it unchanged per ACT and physician order.
-On 2/27/17 between 12:30 p.m. and 4:30 p.m., the Heparin rate stayed the same instead of increasing per ACT and physician order.
-On 2/27/17 at 6:37 p.m. Heparin rate was increased to 0.8 ml/hour instead of 0.7 ml/hour.

Review of facility policy "NICU: Medication Administration" dated 8/7/15 indicated, "It is the responsibility of the Pharmacy and Nursing staff to review all medication orders from physicians and assure they are given as ordered."


29821