HospitalInspections.org

Based on observation, staff interviews, and reviews of patient medical records and hospital administrative records, the hospital failed to develop, maintain, and implement an effective hospital-wide, data-driven quality assessment and performance improvement program regarding medication storage and use in the facility as evidenced by:

1. The hospital failed to identify the medication storage temperatures outside of manufactures's specifications as an opportunity for improvement. This occurred when frozen antibiotics (Azactam, Nafcillin, Penicillin G- potent antibiotics used to treat potentially life-threatening infections) were stored at temperatures required by the manufacturer to maintain their integrity and stability. (Cross reference - A 276)

2. The hospital failed to set priorities for its performance improvement activities that focused on high risk, high volume and problem-prone areas that may affect health outcomes, patient safety and quality of care. (Cross reference - A 285)

The cumulative effects of these systemic problems resulted in the hospital's inability to comply with statutorily mandated Condition of Participation - QAPI.

Based on observation, staff interview, and document review, the hospital failed to identify the medication storage temperatures outside of manufacturers's specifications as an opportunity for improvement. This occurred when frozen antibiotics (Azactam, Nafcillin, Penicillin G) used to treat potentially life-threatening infections were stored for more than two months outside of the temperature range required by the manufacturer to maintain their integrity and stability.

Findings:

On 5/2/11 at 11 a.m., observation of the Freezer #2 at the Eighth Floor Satellite Pharmacy with Pharm 1 and Pharm 2 revealed it contained multiple boxes of Azactam, Nafcillin, and Penicillin G (same manufacturer for all three); each box contained 24 doses. The freezer temperature read -13.1?C (negative 13.1 degrees Celsius).

Review of the manufacturer's storage instructions on the antibiotics' manufacturer labels with Pharm 2 indicated to store at -20?C or below for each product.

On 5/5/11, Pharm 2 provided the temperature reading of medication refrigerators and freezers in the hospital from 2/15/11 to 4/30/11. Review of the temperature readings for Freezer #2 reflect it had temperatures ranging from -15?C to -19?C during this time period. The temperatures never reached -20?C or below.

On 5/5/11 at 11:30 a.m., review of the hospital's policy and procedure on "Medication Storage" with Pharm 2 reflected the freezer temperature was set from -15?C to -20?C; it was established with Pharm 2 that the set temperature range did not meet the requirement for storing the said antibiotics. He acknowledged the finding and said, "You are right! We need to adjust the freezer temperatures to make them colder."

The California Department of Public Health issued an All Facilities Letters (AFL 09-56), in 12/10/09, regarding Medication Safety: Storage of Medications Requiring Refrigeration, the principles of proper medication storage include the following:

"Development and implementation of policies and procedures to ensure safe storage of medications under proper conditions (e.g., temperature, humidity, protection from light and security)...

Storage equipment, e.g., refrigerators, thermometers, carts and cabinets, is maintained in accordance with the standards of practice and/or manufacturer recommendations to ensure proper functioning, which include adherence to pharmaceutical manufacturer requirements for product storage, and an alerting system to quickly identify situations where high and low refrigerator temperature limits are exceeded (this would also be applicable to medical devices used for warming intravenous and irrigating solutions).

All medication storage areas must be periodically inspected by qualified staff in accordance with facility policy, standards of practice, and/or regulatory requirements to ensure that medications are safely and properly stored."

Based on staff interviews, observations, reviews of medical records and reviews of administrative documents, the hospital failed to set priorities for its performance improvement activities that focused on high risk, high volume and problem-prone areas that might affect health outcomes, patient safety, and quality of care. These deficiencies occurred when the hospital failed to:

1. Ensure that 10 of 12 reviewed patients (Patients 3, 5, 6, 7, 8, 9, 10, 11, 12 and 14) received droperidol (used to treat nausea and vomiting) in a manner consistent with the US Food and Drug Administration (FDA) boxed warning. The boxed warning is the strongest warning issued by the FDA and is included in product literature which accompanys each package of droperidol. The boxed warning required a 12 lead ECG (electrocardiogram) to be conducted on each patient before receiving droperidol to determine the patient's QTc interval. The QTc interval measured by an ECG provides information regarding the electrical conduction system (conduction time) in the heart. The boxed warning advises patients with a QTc interval of more than 450 milliseconds for females and more than 440 milliseconds for males should not receive droperidol. The boxed warning also requires an ECG to be performed for two to three hours after completing treatment to monitor for cardiac arrhythmias (irregular rhythms).

2. Ensure pharmacists reviewed all new medication orders before nurses removed the drugs from ADCs (automated drug dispensing cabinets) and administered them to patients. New medication orders were reviewed by pharmacists in order to increase patient safety and confirm the correct choice of medication. This occurred when ADCs were not "profiled" (defined below) and when unnecessary ADC overrides (defined below) may have occurred. Overrides occurred with many dangerous drugs.

3. Determine the disposition of medications when the hospital did not proceed to follow-up on all discrepancies (defined below) in medication counts in their ADCs. This absence of accountability for many dangerous drugs in the ADCs may have resulted in medication errors and diminished staff safety when doses of medications were in excess or missing.

4. Prepare and maintain portable emergency supplies (crash carts) that included a list of medications contained in the supply on the outside. Other failures included: not conducting inspections by pharmacists; lack of accurate and clearly visible lists of medications and the expiration date of the earliest drug to expire on the outside covers of portable emergency containers; not ensuring that all of the medications on the lists were present in the portable containers; and not having a pharmacist seal the emergency medications in the containers. These failures potentially increased the probability that patients would not survive a medical emergency when such medications were needed.

5. Ensure that staff were sufficiently knowledgeable and trained to treat malignant hyperthermia (MH), a potentially fatal medical emergency requiring rapid and effective treatment. The hospital also failed to inform hospital caregivers of the locations of these potential life-saving medications.

6. Ensure that Emergency Department nurses were adequately trained to draw into syringes correct doses of emergency medications for pediatric patients. This occurred when two of three nurses who were tested made errors. Medication errors during medical emergencies potentially placed patients at increased risk to suffer or fail to recover as a result of excessive or inadequate doses of emergency medications.

7. Ensure that nurses accurately administered intravenous (IV) insulin to patients. IV insulin records for two of two reviewed patients demonstrated dosing errors. Administering incorrect doses of insulin potentially increased the probability a patient would suffer from hyperglycemia (high blood sugar) and hypoglycemia (low blood sugar).

8. Ensure that directions for nurses administering potent IV medications were sufficiently detailed in order to prevent them from having to determine how to adjust doses for patients. Instructions for two medications, nitroprusside and nitroglycerin (lower blood pressure), were reviewed. These instructions for nurses provided wide parameters for nurses to determine dosage changes. The parameters did not include dosage adjustment instructions based on patient blood pressure readings or place a lower limit on a patient's blood pressure. These failures increased the unwanted opportunity a patient would suffer from excessive or inadequate doses of medications.


Findings:

1. Droperidol (brand name Inapsine) is a medication with a Food and Drug Administration (FDA: the institution that regulates the approval, labeling, and all regulatory issues regarding medications in the United States) approved use to treat nausea and vomiting. This medication has a boxed warning (also called a "black box" warning) which is the most serious warning required by the Food and Drug Administration to be placed in the product labeling (also called the package insert) for a medication. Boxed warnings document potential problems that can lead to serious injury or death. The boxed warning documents that use of droperidol has been associated with prolongation of the QT interval and/or torsades de pointes (an uncommon but potentially fatal irregular heartbeat).

The QT interval is seen on the ECG of the heart and is the period the lower chambers of the heart undergo electrical stimulation and then "recharge" so to speak. If this period is extended too long, abnormal heart beats may occur including torsades de pointes which can lead to more serious arrhythmias. The boxed warning makes the following points:

i. Cases of QT interval prolongation and/or torsades de points can occur at recommended or below recommended doses.

ii. This has occurred in patients without known risk factors for QT interval prolongation and some cases have been fatal.

iii. All patients should be screened using a 12-lead ECG to check for pre-existing QT prolongation and if present, droperidol should NOT be used (emphasis is in the boxed warning and defined as a QTc more than 440 milliseconds in males or 450 milliseconds in females). The QTc is a "corrected" QT interval as the QT interval can vary with heart rate.

iv. Patients who receive the medication should be monitored via an ECG (i.e. telemetry) prior to and for 2 to 3 hours after receiving the drug.

A review of 12 closed patient electronic medical records revealed the following issues in 10 of those records:

a. On 5/4/11 a review of Patient 3's EMR indicated Patient 3 was a female and her EMR contained a "Cardiac Catheterization Report" dated 3/21/11. The report documented that the patient received 1.25 milligrams (mg) of droperidol intravenously (IV) at 2:44 p.m. on that date. On 5/4/11 at 10:37 a.m. during an interview of RN 19 at that time, she stated that she had worked in the Cardiac Cath Lab for 19 years. She stated that a physician would give a verbal order for the droperidol and the staff nurse would administer it to the patient. She stated in her experience droperidol was commonly used. She stated that all patients in the Cath Lab got a 12 lead ECG before catheterization and staff would use what was available in the EMR: they would not repeat the 12 lead ECG if one was available in the EMR. Patient 3's EMR contained three 12 lead ECGs all done on 3/17/11 at 12:18 p.m. 11:30 p.m. and 11:31 p.m. The QTc intervals reported out on these ECGs were: 402 milliseconds (msec), 491 msec, and 484 msec respectively. Thus the last two ECGs done closest in proximity to the dosing of the droperidol on 3/21/11 were above the threshold for QTc prolongation for females (450 ms) as documented in the boxed warning at which level Patient 3 should not have received this medication according to the boxed warning.

b. On 5/4/11 a review of Patient 6's EMR indicated Patient 6 was a male and his EMR contained a "Cardiac Catheterization Report" dated 4/22/11. The report documented that Patient 6 received 1.25 mg of droperidol IV at 7:56 a.m. on that date. Patient 6's EMR contained two 12 lead ECGs done on 1/27/11 at 1:49 p.m. and 3/17/11 at 3:42 p.m. The QTc intervals reported out on these ECGs were: 461 msec and 459 msec respectively which was above the threshold for QTc prolongation for males (440 msec) as documented in the boxed warning at which level Patient 6 should not have received this medication according the boxed warning. During an interview of RN 20, she stated that these ECGs would have been available for reading by the physician as they had "scan" dates (dates scanned into the EMR by staff) of 2/16/11 and 3/29/11 respectively.

c. On 5/4/11 a review of Patient 7's EMR indicated that Patient 7 was a male and received 1.25 mg of droperidol IV at 5:30 a.m. on 2/5/11. This occurred in the Emergency Department (ED). The patient had been placed on a 3 lead ECG monitor at 2:54 a.m. There was no record of a 12 lead ECG done during the ED visit on 2/5/11 but Patient 7's EMR contained a 12 lead ECG done on 1/7/11 at 9:22 p.m. on which the recorded QTc was 444 msec. This was above the threshold for QTc prolongation for males (440 msec) as documented in the boxed warning at which level Patient 6 should not have received this medication. During an interview of RN 20 at that time she stated that staff had scanned the 1/7/11 ECG into the EMR on 1/8/11 so that it would have been available for the ED physician to review.

d. On 5/5/11 a review of Patient 9's EMR indicated that Patient 9 was a male and received 1.25 mg of droperidol IV on 3/6/11 at 10:30 a.m. and again on 3/7/11 at midnight and 9:23 p.m. while he was a patient in the Cardiac Care Unit (a unit for intensive care of patients with telemetry monitoring and one to one nursing care). Patient 9's EMR contained a 12 lead ECG done on 3/5/11 at 8:55 a.m. on which the recorded QTc interval was 487 msec. This was above the threshold for QTc prolongation for males (440 msec) as documented in the boxed warning at which level Patient 9 should not have received this medication. Staff had scanned this ECG into Patient 9's EMR on 3/5/11 at 9:40 p.m. so it would have been available for review by a physician on 3/6/11 before he received his droperidol.

e. On 5/5/11 a review of Patient 10's EMR indicated that Patient 10 was a female and received 1.25 mg of droperidol IV on 3/11/11 at 3:03 p.m. and again on 3/12/11 at 7:38 p.m. while she was a patient in the Surgical Intensive Care Unit (a unit for intensive care of patients with telemetry monitoring and one to one nursing care). Patient 10's EMR contained a 12 lead ECG done on 3/9/11 at 4:35 p.m. on which the recorded QTc interval was 470 msec. This was above the threshold for QTc prolongation for females (450 msec) as documented in the boxed warning at which level Patient 10 should not have received this medication. Staff had scanned this ECG into Patient 9's EMR on 3/10/11 at 3:14 p.m. so it would have been available for review by a physician on 3/11/11 before Patient 10 received her droperidol.

f. On 5/9/11 a review of Patient 14's EMR indicated that Patient 14 was a female and received 2.5 mg of droperidol IV on 4/2/11 at 9:44 a.m. while she was a patient in the ED. Patient 14's EMR contained three 12 lead ECG done on 3/6/11 at 10:04 a.m., on 3/28/11 at 7:44 a.m. and on 3/31/11 at 5 a.m. on which the recorded QTc intervals were 454 msec, 469 msec, and 454 msec respectively. This was above the threshold for QTc prolongation for females (450 msec) as documented in the boxed warning at which level Patient 14 should not have received this medication. During an interview of RN 20 on this date at 3:45 p.m. she stated that all of these ECGs would have been available for review by the physician before the droperidol was given.

g. On 5/4/11 a review of Patient 5's EMR indicated that Patient 5 received 1.25 mg of droperidol IV at 2:44 p.m. on 3/31/11. The patient had undergone a study of the vasculature in one of her legs even though the study was done in the Cardiac Cath Lab. There was no evidence in the EMR that a physician had ordered and staff had done a 12 lead ECG.

h. On 5/5/11 a review of Patient 8's EMR indicated that Patient 8 received 1.25 mg of droperidol IV on 2/17/11 at 6:10 p.m. while he was recovered in the Post Anesthesia Care Unit (PACU: unit where patients are watched after surgery to make sure they are stable before transfer to a regular hospital floor). There was no record of a 12 lead ECG in the EMR and this was verified by interview at 10:09 a.m. on this date with RN 20.

i. On 5/9/11 a review of Patient 12's EMR indicated that Patient 12 received 2.5 mg of droperidol IV on 2/7/11 at 2:55 a.m. and 2.5 mg IV at 8:05 a.m. on the same date while she was a patient in the ED. The EMR contained no record of a 12 lead ECG done before Patient 12 received her droperidol. Also, there was no evidence (confirmed on 5/10/11 at 12:31 p.m. with RN 20) of any ECG monitoring after Patient 12 received her droperidol.

j. On 5/9/11 a review of Patient 11's EMR indicated that Patient 11 received 2.5 mg of droperidol IV on 2/25/11 at 5:22 p.m. while she was a patient in the ED. Patient 11's EMR contained 12 lead ECGs done on 2/4/11 at 11:59 a.m. and 2/25/11 at 5:11 p.m. on which the recorded QTc intervals were 417 msec and 436 msec respectively which were NOT prolonged according to the boxed warning for droperidol. However, there was no evidence (confirmed on 5/10/11 at 2:21 p.m. with RN 20) of any ECG monitoring after Patient 12 received her droperidol.

On 5/4/11 a review of Policy and Procedure entitled PATIENT CARE STANDARDS IV-55 (Revised 1/4/11) indicated that it had a "Section 1" entitled "Medication Administration". Under "Procedure P. 2" the policy provided the following information and procedures regarding IV (intravenous) droperidol:

"a. The FDA has issued a black box warning about droperidol specific to prolonged QTc ..."

"b. Measure and document the QTc interval prior to administration for patients with known QTc prolongation or for single doses greater than 0.1 mg/kg or 2.5 mg. If greater than 440 msec for males or 450 msec for females or 25% greater than pre-droperidol QTc, do NOT (capital letters in policy) administer and notify the prescriber. The discussion with the prescriber and any follow-up action should be documented in the Notes section of the EMR."

The only way to measure a QT interval (and thus calculate a QTc interval) is to do an ECG. The policy made no provision to ensure that an ECG be done to determine if the QTc interval was prolonged in any given patient if there was no previous record of an ECG in a given patient's clinical record. As written the policy would not prevent the administration of droperidol to a patient with a prolonged QTc interval if, after being informed by staff of the prolonged QTc interval, a physician elected to proceed with the administration of droperidol in such a patient. If it was administered in this case, all that the staff would be required to do would be to document "the discussion with the prescriber and any follow-up action in the Notes section ..." Finally, as written, it would be necessary for staff to determine if the QTc was prolonged based on the patient's gender in order to decide whether or not to give the dose.

On 5/5/11 three nurses were interviewed in the ED. Two out of three nurses did not correctly identify the boxed warning QTc thresholds at which droperidol should not be given.

a. At 8:38 a.m. during an interview of RN 22, she stated that a QTc greater than 380 msec in males and greater than 410 msec in females would be a reason not to give droperidol.

b. At 8:38 a.m. during an interview of RN 23, she stated that a QTc greater than 450 msec in males and 440 msec in females would be a reason not to give droperidol.

Two out of three nurses did not correctly identify the boxed warning QTc thresholds at which droperidol should not be given (the third one did so). However of those two, the one answer was low for both males and females so a patient with a prolonged QTc would not have received droperidol before a physician was contacted. The second nurse had the gender values for the QTc reversed and hence male patients could potentially have received droperidol when they should not have.

On 5/10/11 at 1:22 a.m. during an interview of Pharm 1 he stated that in 2007 the hospital was not following its own protocol and that at that time it was not to be used first line for the treatment of nausea and vomiting. With regard to the boxed warning, he said the drug had the potential to cause QT prolongation. He said that in general he thought that the boxed warning was "excessive". He stated that when he said "I think" he was speaking from the perspective of the Pharmacy and Therapeutics (P&T) Committee (the committee with oversight of the pharmaceutical services in the hospital and which develops policies and procedures establishing safe and effective systems for the procurement, storage, distribution, dispensing and use of drugs in the hospital) and the literature that was available on the subject. He stated that the P&P for droperidol provided on 5/4/11 had been modified "last week" by adding the words "risk factors for" in the sentence "Measure and document the QTc interval prior to administration for patients with known risk factors for QTc prolongation ..."

Pharm 1 produced a copy of the P&T Committee meeting minutes from 12/19/07. A review of the minutes indicated it documented that droperidol may be used first line and administered in doses less than or equal to 2.5 mg with no additional monitoring in patients with no known risk factors for QT prolongation. The prescriber decides what the risk factors are. Pharm 1 stated that these were the current hospital "guidelines" and that the "no known risk factors" part of the minutes is what was used to modify the policy and procedure.

Pharm 1 produced a copy of the P&T Committee meeting minutes from 11/28/07. A review of the minutes indicated it documented that the committee felt that the black box warning with the use of droperidol was "extreme" ..." The committee recommended no ECG monitoring with doses less than or equal to 2.5 mg.

Pharm 1 stated that it was more typical for the hospital to develop medication guidelines and incorporate them into order sets. He said that a guideline was not a policy, it was a recommendation but if specific in nature, they would be expected to be followed. He pointed out that the 1/4/11 PATIENT CARE STANDARDS IV-55 was not congruent with the above droperidol guidelines. He stated the policy was developed after the "guidelines" and the two should have been consistent but they were not. He could not say during the interview which of these, a guideline or a policy and procedure was the determinate factor for action. He didn't know.

Pharm 1 stated he felt that an increased QTc interval was a risk factor with the use of droperidol and therefore would be a reason to contact the prescriber per the PATIENT CARE STANDARDS IV-55.

2. Automated drug dispensing cabinets:
a. On 5/2/11 at 10:00 a.m., Pharm (pharmacist) 1 and Pharm 2 said hospital automated drug dispensing cabinets (ADCs) were not "profiled" in medical procedural areas such as the gastrointestinal laboratory and the cardiac catheterization laboratory, Labor and Delivery, Post Anesthesia Recovery (PACU), and the Emergency Department. The ADCs were electronically connected to the hospital computer system. Profiled ADCs were those which required a pharmacist to review prescribers' new medication orders and approve them electronically before a nurse could remove the ordered medication from the ADC. Prescribers (for example, physicians) entered new medication orders directly into the hospital computer and pharmacists were able to view them on the computer screen. Medications obtained from ADCs which were not profiled did not require a pharmacist to review the new medication orders before removal. Pharmacists reviewed new medication orders for several safety and efficacy (effective drug for the disease) reasons which included: ensuring the correct medication and dose was ordered and to avoid unwanted drug interactions, adverse drug effects (side effects), drug allergies, and medication errors. Medications removed for patients from ADCs which were not profiled did not benefit from these safety and efficacy checks.

Examination of the hospital's medication error reduction plan ("UC Davis Medical Center Medication-Related Error Reduction Plan December 19, 2001") indicated on page 9 that, "Nurses do not remove medication from PYXIS (ADC) until a pharmacist has reviewed the order." Under" Comment" on this listing was the conclusion, "Done, exception is emergency situations."

b. On 5/2/11 at 10:30 a.m., Pharm 1 said the "override" function was used for about one percent of the medications removed from ADCs. Overrides occurred on profiled ADCs when medications were removed before a pharmacist had reviewed the new medication orders. Overrides prevented pharmacists from reviewing new medication orders for safety and efficacy checks before patients received the medications.

A review of the hospital's report on overrides for January, 2011 ("Rx Auditor, Override Summary Analysis, January 2011") demonstrated that the number of overrides were 1,980 which were 1.05 percent of 188,492"dispense events". A dispense event occurred each time a nurse removed medication from an ADC, usually for a particular patient. The February, 2011 report listed 2,036 overrides and the March 2011 report listed 1,708 overrides.

The hospital provided a list of 93 medications which could be removed from all ADCs and a separate list of 53 medications which could be removed on override only from designated ADCs. Examples of medications on these lists included muscle paralyzing agents (succinylcholine, vercuronium), narcotic type analgesics (fentanyl, hydromorphone, morphine), sedatives (lorazepam, diazepam, phenobarbital), medications to increase blood pressure, decrease blood pressure, and medications to increase the heart rate. Many of the medications on the lists could have been lethal (could kill you) if they were administered in overdose or to the wrong patient.

Hospital policy and procedures ("Automated Medication Dispensing Machines, revised 5/13/09") read, "these [override] medications are approved based on consideration of urgency of need, time to onset of action, potential dangers of administration and lack of need for dosage adjustment in renal/hepatic dysfunction." Some of the reasons listed in the computer for medications removed on override ("Override Detail Response Analysis, March, 2011") included: "Procedural use; RN (registered nurse)/RT (respiratory therapist) clinical evaluation of patient status; to relieve pain; MD states one dose needed urgently; no allergy to this medicine; to flush an IV (intravenous) catheter". Many of the overrides did not have a reason for removal and others were nonsensensical (example:"no-no action").

The American Society of Health-Systems Pharmacists (ASHP) have established the standards of practice for ADCs ("ASHP Guidelines on the Safe Use of Automated Medication Storage and Distribution Devices: Am J Health-Syst Pharm. 1998; 55:1403-07"). The ASHP guidelines indicate that the automated system or device should ..."Provide for prospective, timely review of medication orders by a pharmacist at all appropriate decision points in the medication-use process, especially before administration of the first dose; and provide for the independent interpretation of the medication order by a pharmacist and a nurse." ASHP also points out that"Clearly stated organizational policies should be developed that limit access to medications before orders have been reviewed and approved by a pharmacist. Access to medications should be limited to the following cases: 1. The order has been reviewed and approved by a pharmacist. 2. The drug product has been approved by a multidisciplinary committee ... who agree that it has minimal risk for misadventures. 3. There is a clinically urgent need for the medication that outweighs the potential risk. ..". Also, "Provision should be made for the retrospective review and reconciliation by a pharmacist of orders that were initiated without a pharmacist's review and approval." At 8:00 a.m. on 5/3/11, Pharm 1 said the hospital did not follow-up on overrides to determine whether the medications were removed according to hospital policy. He did, however, say that pharmacists retrospectively (after the override) followed-up on overrides to ensure that a prescriber's order had been entered into the computer.

3. On 5/2/11 at 10:30 a.m., Pharm 1 said the hospital did not follow-up on all "discrepancies" from ADCs. He said the hospital only followed-up to determine the disposition (what happened to the medication) of controlled substances discrepancies. Controlled substances included narcotic analgesics (for example, morphine) and sedative type medications (for example, phenobarbital and Valium). A discrepancy occurred when the number of doses of a medication in the ADC was discovered to be different from the number of doses listed in the computer attached to the ADC. The number of doses placed in the ADC were listed in the attached computer when the medication drawer or pocket in the ADC was filled. For example, discrepancies could occur when a nurse indicated in the computer that she removed one dose but actually removed two or more. Also, a discrepancy could occur if a nurse listed that she removed one but changed her/his mind and did not remove it. Potential problems attributable to discrepancies included drug diversion (employee taking medication for own use), administration of too many or too few doses, and other medication errors such as giving the medication to the wrong patient.

A review of the hospital's reports on ADC discrepancies from 2/1/11 through 4/28/11 showed that there were approximately 400 controlled substances discrepancies and approximately 530 discrepancies for other potentially dangerous medications. The controlled substances discrepancies listed the reasons for the discrepancies and the resolution. The non-controlled substances discrepancies did not list reasons for discrepancies nor resolutions.

The hospital's policy and procedures for ADCs ("Automated Medication Dispensing Machines, revised 5/13/09") did not address the issue of dealing with or resolving discrepancies for medications which were not controlled substances. Examples of these discrepancies included dangerous medications such as succinylcholine and rocuronium (cause muscle paralysis), atropine (increases the heart rate), propofol and dexmedetomidine (sedatives), nitroprusside (lowers blood pressure), nicotine patches, quetiapine (antipsychotic for mental illness), antibiotics, and heparin and warfarin (decrease blood clotting).

ASHP ("ASHP Guidelines on the Safe Use of Automated Medication Storage and Distribution Devices: Am J Health-Syst Pharm. 1998; 55:1403-07") indicates that "The organization should have a written plan for the monitoring and surveillance of medications accessed through automated systems. ... The plan should include identification of the data to be captured and the reports generated that are used to monitor medication use (data and reports may vary by drug categories and requirements for control and accountability). ... A description of the process for investigating trends in discrepancies and assigning responsibility for this."

4. On 5/2/11 at 10:45 a.m., transport kits were observed in the pharmacy. Pharm 2 said the portable transport kits which contained medications were filled in the pharmacy. He said they were emergency kits but did not contain lists of medications on the outside cover of the container. Pharm 2 said the lists were on the inside of the container. The contents of the container shall be listed on the outside cover and shall include the earliest expiration date of any drugs within.

On 5/2/11 at 1:30 p.m., RN (registered nurse) 9 in the Medical Intensive Care Unit (MICU) said she did not know there was a list of medications on the outside cover of their portable emergency cart.

On 5/2/11 at 2:35 p.m., RN (registered nurse) 1 in the Post Anesthesia Recovery Unit (PACU) area said the pediatric and adult airway boxes located there were portable emergency containers with medications inside. The adult box was opened and the following were examples of medications which were observed inside: lidocaine oral solution (for oral pain), vercuronium and succinylcholine (for muscle paralysis), and atropine (increases heart rate). There were 17 medications inside. The container did not have a list of medications and did not have an expiration date of the first medication to expire on the outside cover. MD (physician) 1 said she stocked (placed medications inside) the adult airway box and sealed it with a yellow plastic lock from a supply which was located in the PACU.

On 5/2/11 at 2:40 p.m., an anesthesia code box, which was located in the PACU, was inspected with RN 1 and RN 2. It was a portable emergency container with 24 medications inside. It was sealed with a yellow plastic lock and MD 1 said she placed medications inside the box and sealed the container. MD 1 said it was the hospital's practice for the anesthesiology resident to prepare the anesthesia code box with medications and equipment. The box did not contain a list of medications and the expiration date of the first drug to expire on the outside cover. It also contained the expired medications, naloxone (to reverse narcotic analgesics such as morphine) and rocuronium (for muscle paralysis).

On 5/3/11 at 10:50 a.m., an adult airway emergency box in the pediatric emergency department area was examined with RN 2 and RN 3. RN 3 said the box was delivered by Central Processing and a third year medical resident sealed the box. The emergency box contained medications and did not have a list of medications and the expiration date of the first drug to expire on the outside cover.

On 5/4/11 at 9:15 a.m., naloxone and flumazenil (reverses some sedatives) were found in the Burn Unit in a cabinet over a sink. RN 4, in the presence of RN 2, said "pharmacy doesn't want us opening the code (portable emergency) cart for relatively minor items." He said nurses removed medications such as atropine, epinephrine, and lidocaine from the ADCs when they transported patients. RN 4 indicated that these were medications which might be necessary to treat a patient during transport if a medical emergency occurred. He said nurses obtained these medications from the ADC on the override function. RN 4 said nurses placed the medications in a zip-lock bag and carried them in their pockets during transport. He also was unable to locate the list of medications on the portable emergency cart. The list was not visible as it was located on the rear of the cart facing the wall.

On 5/4/11 at 11:05 a.m., the portable emergency cart containing medications in the Neonatal Intensive Care Unit was examined with RN 2. It contained one-250 ml (milliliter) container of 5% dextrose (sugar) solution and one-500 ml container of 10% dextrose solution. The list on the outside cover of the cart indicated that there should have been two of each in the cart. There also were two-500 ml containers of normal saline (salt solution) and one-120 ml container of povidone-iodine (antibacterial, antifungal solution) in drawer five which were not on the list of medications on the outside cover of the cart. At 11:45 a.m. Pharm 7 said "Central Processing takes care of the list of medications on the portable emergency carts".

The following include a sample of the dates nursing unit inspections were conducted. The inspections also included portable emergency carts containing medications which were checked to ensure they contained all of the medications on the list on the outside cover, the expiration date of the earliest drug to expire, no expired medications, and were sealed by a pharmacist.

Vascular: 4/8/11, 3/31/11, 2/25/11, 1/?/11
Obstetrics/Gynecology: 4/?/11, 3/?/11, 2/?/11, 1/?/11
Ophthalmology: 3/?/11, 2/?/11, 1/?/11
(? = no day listed)

On 5/5/11 at 8:00 a.m., Pharm 1 said emergency carts containing medications were checked mont

Based on observations, staff interview and medical record review the hospital failed to administer medications in accordance with the physician orders and the hospital policies and procedures (P&P) for 2 of 30 sampled patients reviewed (Patient 22 and 30).

Findings:

1. The electronic medical record (EMR) for Patient 22 was reviewed on the ENT (Ear/Nose throat) postoperative unit on 5/9/11 at 11:25 a.m. with RN 10 (Registered Nurse assigned to care for Patient 22). The History and Physical showed that Patient 22 was admitted to the hospital on 5/8/11 for surgery after suffering multiple fractures from a motor vehicle accident. The nursing record showed that Patient 22 arrived to the postoperative unit on 5/8/11 at 8 p.m.

A review of the physician's orders showed that orders included an order on 5/8/11 at 8 p.m. for Vancomycin (antibiotic) 1 Gram intravenously (via infusion over 60 minutes) every 12 hours, to start now, for surgical prophylaxis, end 5/9/11.

A review of Medication Administration Record (MAR) showed that, Vancomycin 1 Gram scheduled for 5/9/11 at 8 a.m. was not administered as of the time of review at 11:25 a.m. RN 10 stated that this dose was overdue because the Vancomycin was not available at 8 a.m. RN 10 was asked if the pharmacy was aware of the order and of the missing dose and stated that the order was forwarded to the pharmacy via EMR and was acknowledged by the pharmacy. It should have been prepared and brought in to the unit by a pharmacist.

At 11:55 a.m. a staff was observed handing the dose of Vancomycin to RN 10, who then went into Patient 22's room and initiated the infusion of Vancomycin at noon.

On 5/9/11 at 12:05 p.m. the unit pharmacist was interviewed regarding the availability of the ordered Vancomycin. The pharmacist stated that the dose was in the refrigerator all morning located in the unit medication refrigerator room. Apparently RN 10 did not see it because it was folded and did not alert the pharmacy that the dose was missing.

A review of P&P titled Medication Administration IV-55 (revised 1/4/11) indicated, "D. Routine medications should be administered within 30 minutes of the designated time." The "procedure" section indicated that after the pharmacist reviews and validates the medication order, the registered nurse will access the medication from locations on the unit that include the medication refrigerator. The P&P indicated that the registered nurse may contact the pharmacy directly to review an order 24 hours a day, 7 days a week.

The Vancomycin for Patient 22 was administered 4 hours after the ordered time and 3.5 hours late as per the facility P&P.


2. Patient 30 was admitted to the Cardio-Thoracic unit on 5/8/11. Patient 30's physician ordered Tiotropium Bromide (Spiriva)18 mcg (microgram) capsules with inhalation device to be given starting 5/9/11 at 9:00 a.m., for daily treatments in the morning. A review of the Medication Administration Record (MAR) was conducted on 5/10/11 by the Department and it was noted on the MAR that the medication was not given by the Respiratory Therapist (RT). Interviews with the Assistant Nursing Director and a Pharmacist confirmed that the medication had not been given by the appropriate staff (Respiratory Therapist).

A review of the Medication Detail portion of the MAR showed that the RT noted "No handi-haler available, Rx (pharmacy) notified." Further noted on the detail sheet was the notation under administration instructions: Must use with HandiHaler device.

The RT responsible for this medication was interviewed on 5/10/11 about this missed medication. The RT confirmed that the medication was not administered because the inhalation device was not available with the medication. The RT stated that the pharmacy was notified but that she did not follow-up later with the pharmacy or prescribing physician.

An interview was conducted with the Respiratory Therapy Department Manager (RT Mgr) on 5/10/11 at 12:05 p.m. The RT Mgr stated that the medication that was missed is on a 24 hour time frame and that the RTs would not want to give it several hours later because it would push back the rest of the timing for this medication and they try to set a schedule that patients can follow at home. When asked if the RT is responsible for informing the physician if a medication is missed, the RT Mgr stated that the RT did not notify the physician that the inhalation device was not available thus missing the administration of the medication. The RT Mgr stated RTs are not required to do the Medication Management portion of online training that the Nursing staff must pass.

On 5/10/11, at 1:20 p.m., an interview was conducted with the Assistant Nursing Director regarding this incident and he confirmed that this was a medication error according to the hospital Policy and Procedure titled Medication Administration by Non-Licensed Personnel-IV-55, revised 1/4/11. Item L in this policy identifies that "For errors that reach the patient, the responsible medical service should be notified immediately per UCDMC Policy 1633." Also noted under item O is "In EMR (Electronic Medical Record), if an ordered dose of medication is not given, enter the reason for not giving the medication and the name of the physician notified."

Based on observations made in patient care areas, reviews of patient medical records, reviews of administrative documents, and staff interviews, the hospital failed to ensure that the pharmaceutical services met the needs of the hospital patients as evidenced by the failure of the hospital to:

1. a) Ensure its staff carried out the appropriate procedure (to dispatch the alarm), for when the investigation drug refrigerator (REF 1) temperature went out of range, in accordance with the hospital's policy and procedure. As a result, REF 1 was out of range (below established temperature range) for 10 days and b) Ensure that frozen medications (Azactam, Nafcillin, Penicillin G- potent antibiotics used to treat potentially life-threatening infections) were stored at the temperature range in accordance with manufacturer's instructions in order to maintain medication integrity and stability. (Cross reference - A491)

2. Ensure that the operations of the Pharmacy Department provided safe and effective medications for administration to patients. (Cross reference - A492)

3. Ensure patient safety when drugs and biologicals were not controlled and distributed in accordance with State regulations; the hsopital failed to develop and implment policies and procedures for safe and effective systems fordistribution, dispending, and use of medications and biologicals (Cross reference A500)

4. Remove expired and other unusable medications from patient care areas. Administration of such medications could have produced unexpected (for example, reduced) effects when administered to patients. Unexpected effects of medications could have impaired and/or delayed a patient's recovery. (Cross reference A505)

5. The hospital failed to maintain appropriate temperature controls in one refrigeration unit, the investigational drug refrigerator (REF 1) in the pharmacy, for 10 days in April 2011. The failure to main refrigeration unit may result in compromised medication integrity and stability, which could result in unsafe and/or ineffective medications for patients. (Cross reference A726)

The cumulative effects of these systemic problems resulted in the hospital's inability to comply with statutorily mandated Condition of Participation - Pharmaceutical Services.

Based on observation, staff interviews, and document reviews, the hospital failed to:

1. Ensure its staff carried out the appropriate procedure (to dispatch the alarm), for when the investigation drug refrigerator (REF 1) temperature went out of range, in accordance with the hospital's policy and procedure. As a result, REF 1 was out of range (below established temperature range) for 10 days. The hospital lacked evidence that it instituted changes to its current monitoring system to prevent recurrence of of the incident.

2. Ensure that frozen medications (Azactam, Nafcillin, Penicillin G- potent antibiotics used to treat potentially life-threatening infections) were stored at the temperature range in accordance with manufacturer's instructions in order to maintain medication integrity and stability.

These failures had the potential for unsafe and ineffective systems for the storage, distribution, and use of medications in the hospital, which could result in unnecessary adverse patient outcomes.

Findings:

1. On 5/2/11 at 10 a.m., during a tour to the Central Pharmacy, the Director of Pharmacy (Pharm 1) said the hospital had a electronic monitoring system (called the Metasys) for medication refrigerators and freezers in the hospital's pharmacies. He said it was the responsibility of the Operations & Maintenance (POM) to maintain and monitor Metasys-operated refrigeration units in the hospital's pharmacies. He explained that when the temperature went out of range, the system would set off an alarm. The POM operator would call the pharmacist or send a personnel to investigate and revolve the issue.

On 5/5/11 at 8:30 a.m., the Pharmacy Assistant Manager (Pharm 2) provided the temperature readings of the medication refrigerators and freezers in the hospital's pharmacies from 2/15/11 to 4/30/11.

On 5/5/11 at 11:30 a.m., Pharm 2 said the hospital set the temperature parameters for all medication refrigerators at 2?C to 8?C. Review of the REF 1 with Pharm 2 revealed the temperatures were continuously out of range for 10 days from 4/11/11 to 4/21/11, ranging from 1.38?C to as low as -1.47?C (below freezing, which is 0?C). When asked if he was aware of the out-of-range temperature for this time period, Pharm 2 said, "I don't know."

On 5/5/11 at 12 p.m., a visit to the Central Pharmacy was conducted with Pharm 2 and a Staff Pharmacist (Pharm 6). Observation of REF 1 reflected it contained various investigational drugs, each in bulk quantities (the hospital had ongoing investigational studies). Pharm 6 said the temperature requirement for these investigational drugs was from 2 to 8?C. REF 1 had three different types of temperature monitoring devices in it during this visit.

When asked why there were more than one type of monitoring systems in REF 1, Pharm 6 said REF 1 had a temperature issue some time in March 2011, at which time it was put out of service, and all medications were quarantined. He explained he had to have other monitoring devices in it "just to be sure." Pharm 6 said he continually paid attention to its temperatures since it was put back in service (on 4/7/11) and became aware of out-of-range temperatures on 4/20/11. At that time he immediately notified POM. Pharm 6 explained, in the 4/20/11 e-mail communication with the POM Manager (POM MGR), he was told the temperature went out of range on 4/11/11 at 4:52 p.m. and the system did set off an alarm, however, the POM operator failed to dispatch the alarm by failing to send a personnel to investigate the affected site.

On 5/5/11 at 2 p.m., the POM MGR was interviewed in the presence of Pharm 2. He confirmed Pharm 6's statement. He said, per hospital's policy and procedure, it was the responsibility the POM to operate, monitor, and respond to alarms, of the hospital's automation system including pharmacy's refrigerators. He said "the operator failed to dispatch the alarm" on 4/11/11, and therefore the system "stayed in alarm". However, it would not stay active as a "pop-up" on the computer screen, but remained in the "stay alarm" functionality, which was not visible to the operator, therefore it was not acted upon until Pharm 6's notification. As a result, the affected refrigerator (REF 1) stayed out of range for 10 days, from 4/11/11 to 4/21/11.

POM MGR explained that when comparing the temperature readings by Metasys versus those of the Pharm 6's other monitoring devices, the temperatures differed by 1?C; he figured it was the calibration issue that explained the difference in temperature readings. Consequently, he sent out a technician to calibrate the sensor of REF 1.

When asked what had been done to prevent recurrence of the same type incident, POM MGR said he "spoke to the operator about the appropriate procedures by sending a dispatch and to actively look for the 'stay alarm' warnings." He said he will also educate other operators about the same procedures. However, there had not been any changes in the way the system was set up to help operators visibly identify the "stay alarm" warnings should the same incident happen again. There was also no documented evidence the hospital looked into the calibration issue that could have affected other Metasys-monitored refrigeration units in the hospital.

2. On 5/2/11 at 11 a.m., a tour was conducted at the Eighth Floor Satellite Pharmacy with Pharm 1 and Pharm 2. Observation of the Freezer #2 revealed it contained multiple boxes of Azactam, Nafcillin, and Penicillin G; each box contained 24 doses. The freezer temperature read -13.1?C (negative 13.1 degrees Celsius).

Review of the manufacturer's storage instructions on the antibiotics' manufacturer labels with Pharm 2 indicated to store at -20?C or below for each product.

On 5/5/11, review of the temperature readings from 2/15/11 to 4/30/11 reflected the medication Freezer #2 had temperatures ranging from -15?C to -19?C during this time period. The temperatures never reached -20?C or below.

On 5/5/11 at 11:30 a.m., review of the hospital's policy and procedure on "Medication Storage" with Pharm 2 reflected the freezer temperature was set from -15?C to -20?C; it was established with Pharm 2 that the set temperature range did not meet the requirement for storing the said antibiotics. He acknowledged the finding and said, "You are right! We need to adjust the freezer temperatures to make them colder."

Based on observations, staff interviews, and reviews of administrative documents, the hospital's Pharmacy Director failed to ensure that the operations of the Pharmacy Department provided safe and effective medications for administration to patients. These deficiencies included:

1. The failure to ensure pharmacists reviewed all new medication orders before nurses removed the drugs from ADCs (automated drug dispensing cabinets) and administered them to patients. New medication orders were reviewed by pharmacists in order to increase patient safety and confirm the correct choice of medication. This occurred when ADCs were not "profiled" (defined below) and when unnecessary ADC overrides (defined below) may have occurred. Overrides occurred with many dangerous drugs.

The failure to determine the disposition of medications when the hospital did not proceed to follow-up on all discrepancies (defined below) in medication counts in their ADCs. This absence of accountability for many dangerous drugs in the ADCs may have resulted in medication errors and diminished staff safety when doses of medications in ADCs were in excess or were missing.

2. The failure to have pharmacists conduct unit inspections. Inspections of medication storage in each unit by pharmacists can provide medication use assessment and prevent the hospital from unsafe practices such as administering expired medications and medications which have been stored at incorrect temperatures or are otherwise deteriorated.

3. The failure to prepare and maintain portable emergency supplies. Failures included: not conducting inspections by pharmacists; lack of accurate and clearly visible lists of medications and the expiration date of the earliest drug to expire on the outside covers of portable emergency containers; not ensuring that all of the medications on the lists were present in the portable containers; and not having a pharmacist seal the emergency medications in the containers. These failures potentially increased the probability that patients would not survive a medical emergency when such medications were needed.

4. The failure to remove expired and other unusable medications from patient care areas. Administration of such medications could have produced unexpected (for example, reduced) effects when administered to patients. Unexpected effects of medications could have impaired and delayed patients' recoveries.

Findings:

1. Automated drug dispensing cabinets:
a. On 5/2/11 at 10:00 a.m., Pharm (pharmacist) 1 and Pharm 2 said hospital automated drug dispensing cabinets were not "profiled" in medical procedural areas such as the gastrointestinal laboratory and the cardiac catheterization laboratory, Labor and Delivery, Post Anesthesia Recovery (PACU), and the Emergency Department. The ADCs were electronically connected to the hospital computer system. Profiled ADCs were those which required a pharmacist to review prescribers' new medication orders and approve them electronically before a nurse could remove the ordered medication from the ADC. Prescribers (for example, physicians) entered new medication orders directly into the hospital computer and pharmacists were able to view them on the computer screen. Medications obtained from ADCs which were not profiled did not require a pharmacist to review the new medication orders before removal. Pharmacists reviewed new medication orders for several safety and efficacy (effective drug for the disease) reasons which included: ensuring the correct medication and dose was ordered and to avoid unwanted drug interactions, adverse drug effects (side effects), drug allergies, and medication errors. Medications removed for patients from ADCs which were not profiled did not benefit from these safety and efficacy checks.

Examination of the hospital's medication error reduction plan ("UC Davis Medical Center Medication-Related Error Reduction Plan December 19, 2001") indicated on page 9 that,"Nurses do not remove medication from PYXIS (ADC) until a pharmacist has reviewed the order. " Under"Comment" on this listing was the conclusion, "Done, exception is emergency situations."

b. On 5/2/11 at 10:30 a.m., Pharm 1 said the "override" function was used for about one percent of the medications removed from ADCs. Overrides occurred on profiled ADCs when medications were removed before a pharmacist had reviewed the new medication orders. Overrides prevented pharmacists from reviewing new medication orders for safety and efficacy before patients received the medications.

A review of the hospital's report on overrides for January, 2011 ("Rx Auditor, Override Summary Analysis, January 2011") demonstrated that the number of overrides were 1,980 which were 1.05 percent of 188,492 "dispense events". A dispense event occurred each time a nurse removed medication from an ADC, usually for a particular patient. The February, 2011 report listed 2,036 overrides and the March 2011 report listed 1,708 overrides.

The hospital provided a list of 93 medications which could be removed from all ADCs and a separate list of 53 medications which could be removed on override only from designated ADCs. Examples of medications on these lists included muscle paralyzing agents (succinylcholine, vecuronium), narcotic type analgesics (fentanyl, hydromorphone, morphine), sedatives (lorazepam, diazepam, phenobarbital), medications to increase and decrease blood pressure, and medications to increase the heart rate. Many of the medications on the lists could have been lethal (could kill you) if they were administered in overdose or to the wrong patient.

Hospital policy and procedures ("Automated Medication Dispensing Machines, revised 5/13/09") read, "these [override] medications are approved based on consideration of urgency of need, time to onset of action, potential dangers of administration and lack of need for dosage adjustment in renal/hepatic dysfunction." Some of the reasons listed in the computer for medications removed on override ("Override Detail Response Analysis, March, 2011") included: "Procedural use; RN (registered nurse)/RT (respiratory therapist) clinical evaluation of patient status; to relieve pain; MD states one dose needed urgently; no allergy to this medicine; to flush an IV (intravenous) catheter". Many of the overrides did not have a reason for removal and others were nonsensensical (example: "no-no action").

The American Society of Health-Systems Pharmacists (ASHP) have established the standards of practice for ADCs ("ASHP Guidelines on the Safe Use of Automated Medication Storage and Distribution Devices: Am J Health-Syst Pharm. 1998; 55:1403-07"). The ASHP guidelines indicate that the automated system or device should ..."Provide for prospective, timely review of medication orders by a pharmacist at all appropriate decision points in the medication-use process, especially before administration of the first dose; and provide for the independent interpretation of the medication order by a pharmacist and a nurse." ASHP also indicates that "Clearly stated organizational policies should be developed that limit access to medications before orders have been reviewed and approved by a pharmacist. Access to medications should be limited to the following cases: 1. The order has been reviewed and approved by a pharmacist. 2. The drug product has been approved by a multidisciplinary committee ... who agree that it has minimal risk for misadventures. 3. There is a clinically urgent need for the medication that outweighs the potential risk ..." Also, "Provision should be made for the retrospective review and reconciliation by a pharmacist of orders that were initiated without a pharmacist's review and approval." At 8:00 a.m. on 5/3/11, Pharm 1 said the hospital did not follow-up on overrides to determine whether the medications were removed according to hospital policy. He did, however, say that pharmacists retrospectively (after the override) followed-up on overrides to ensure that a prescriber's order had been entered into the computer.

c. On 5/2/11 at 10:30 a.m., Pharm 1 said the hospital did not follow-up on all "discrepancies" from ADCs. He said the hospital only followed-up to determine the disposition (what happened to the medication) of controlled substances discrepancies. Controlled substances included narcotic analgesics (for example, morphine) and sedative type medications (for example, phenobarbital and Valium). A discrepancy occurred when the number of doses of a medication in the ADC was discovered to be different from the number of doses listed in the computer attached to the ADC. The number of doses placed in the ADC were listed in the attached computer when the medication drawer or pocket in the ADC was filled. Examples of discrepancies included: when a nurse indicated in the computer that she removed one dose but actually removed two or more; when a nurse listed in the computer that she removed one dose but changed her/his mind and did not remove it. Potential problems attributable to discrepancies included drug diversion (employee taking medication for own use), administration of too many or too few doses, and other medication errors such as giving the medication to the wrong patient.

A review of the hospital's reports on ADC discrepancies from 2/1/11 through 4/28/11 showed that there were approximately 400 controlled substances discrepancies and approximately 530 discrepancies for other potentially dangerous medications. The controlled substances discrepancies listed the reasons for the discrepancies and the resolution. The non-controlled substances discrepancies did not list reasons for discrepancies nor resolutions.

The hospital's policy and procedures for ADCs ("Automated Medication Dispensing Machines, revised 5/13/09") did not address the issue of dealing with or resolving discrepancies for medications which were not controlled substances. Examples of these discrepancies included dangerous medications such as succinylcholine and rocuronium (cause muscle paralysis), atropine (increases the heart rate), propofol and dexmedetomidine (sedatives), nitroprusside (lowers blood pressure), nicotine patches, quetiapine (antipsychotic for mental illness), antibiotics, and heparin and warfarin (decrease blood clotting).

ASHP ("ASHP Guidelines on the Safe Use of Automated Medication Storage and Distribution Devices: Am J Health-Syst Pharm. 1998; 55:1403-07") indicates that "The organization should have a written plan for the monitoring and surveillance of medications accessed through automated systems. ... The plan should include identification of the data to be captured and the reports generated that are used to monitor medication use (data and reports may vary by drug categories and requirements for control and accountability). ... A description of the process for investigating trends in discrepancies and assigning responsibility for this."

2. At 10:20 a.m. on 5/2/11, Pharm 1 said Pharm Tech (pharmacy technician) 1 conducted all of the hospital nursing unit inspections with a pharmacist. Inspections of medication storage in each nursing unit by pharmacists provide a method for assessment of medcation use and can prevent the hospital from unsafe practices such as administering expired medications and medications which have been stored at incorrect temperatures or are otherwise deteriorated. Pharmacists are educated to identify potential problems attributable to medications during unit inspections; pharmacy technicians do not have this breadth of knowledge. There is no provision for anyone other than a pharmacist to conduct the unit inspections, and that is directed by State requirement. The California Code of Regulations (CCR) Title 22, 70263(q)(10) read: "Drugs maintained on the nursing unit shall be inspected at least monthly by a pharmacist."

3. On 5/2/11 at 10:45 a.m., transport kits were observed in the pharmacy. Pharm 2 said the portable transport kits which contained medications were filled in the pharmacy. He said they were emergency kits but did not contain lists of medications on the outside cover of the container. He said the lists were on the inside of the container.

On 5/2/11 at 1:30 p.m., RN (registered nurse) 9 in the Medical Intensive Care Unit (MICU) said she did not know there was a list of medications on the outside cover of portable emergency carts.

On 5/2/11 at 2:35 p.m., RN 1 in the Post Anesthesia Recovery Unit (PACU) area said the pediatric and adult airway boxes located there were portable emergency containers with medications. The adult box was opened and the following were examples of medications which were observed inside: lidocaine oral solution (for oral pain), vecuronium and succinylcholine (for muscle paralysis), and atropine (increases heart rate). There were 17 medications inside. The container did not have a list of medications and did not have an expiration date of the first medication to expire on the outside cover. MD (physician) 1 said she stocked (placed medications inside) the adult airway box and sealed it with a yellow plastic lock from a supply which was located in the PACU.

On 5/2/11 at 2:40 p.m., an anesthesia code box, which was located in the PACU, was inspected with RN 1 and RN 2. It was a portable emergency container with 24 medications inside. It was sealed with a yellow plastic lock and MD 1 said she placed medications inside the box and sealed the container. MD 1 said it was the hospital's practice for the anesthesiology resident to prepare the anesthesia code box with medications and equipment. The box did not contain a list of medications and the expiration date of the first drug to expire on the outside cover. It also contained the expired medications, naloxone (to reverse narcotic analgesics such as morphine) and rocuronium (for muscle paralysis).

On 5/3/11 at 10:50 a.m., an adult airway emergency box in the Pediatric Emergency Department area was examined with RN 2 and RN 3. RN 3 said the box was delivered by Central Processing and a third year medical resident sealed the box. The emergency box contained medications and did not have a list of medications and the expiration date of the first drug to expire on the outside cover.

To ensure availability of emergency medications and ease of access, it is a standard of practice to have the list of the content of the emergency supplies on the outside of the box. That is also dictated by state regulations. California Code of Regulation (CCR), Title 22, section 70263 (f)(2) required that "The emergency drug supply shall be stored in a clearly marked portable container which is sealed by the pharmacist in such a manner that a seal must be broken to gain access to the drugs. The contents of the container shall be listed on the outside cover and shall include the earliest expiration date of any drugs within."

On 5/4/11 at 9:15 a.m., naloxone and flumazenil (reverses some sedatives) were found in the Burn Unit in a cabinet over a sink. RN 4, in the presence of RN 2, said "pharmacy doesn't want us opening the code (portable emergency) cart for relatively minor items." He said nurses removed medications such as atropine, epinephrine (increases blood pressure and heart rate), and lidocaine (for irregular heart rhythms) from the ADCs when they transported patients. RN 4 indicated that these were medications which might be necessary to treat a patient during transport if a medical emergency occurred. He said nurses obtained these medications from the ADC on the override function. RN 4 said nurses placed the medications in a zip-lock bag and carried them in their pockets during transport. He also was unable to locate the list of medications on the portable emergency cart located in the Burn Unit. The list was not visible as it was located on the rear of the cart facing the wall.

On 5/4/11 at 11:05 a.m., the portable emergency cart containing medications in the Neonatal Intensive Care Unit was examined with RN 2. It contained one-250 ml (milliliter) container of 5% dextrose (sugar) solution and one-500 ml container of 10% dextrose solution. The list on the outside cover of the cart indicated that there should have been two of each in the cart. There also were two-500 ml containers of normal saline (salt solution) and one-120 ml container of povidone-iodine (antibacterial, antifungal solution) in drawer five which were not on the list of medications on the outside cover of the cart. At 11:45 a.m. Pharm 7 said "Central Processing takes care of the list of medications on the portable emergency carts".

The following include a sample of nursing unit inspections and the dates they were conducted. The inspections also included portable emergency carts containing medications which were checked to ensure they contained all of the medications on the list on the outside cover, the expiration date of the earliest drug to expire, no expired medications, and were sealed by a pharmacist.

Vascular: 4/8/11, 3/31/11, 2/25/11, 1/?/11
Obstetrics/Gynecology: 4/?/11, 3/?/11, 2/?/11, 1/?/11
Ophthalmology: 3/?/11, 2/?/11, 1/?/11
(? = no day listed)

Hospital policy and procedures, "Maintenance and Inspection of Crash (Code) Cart Medications", read, "Pharmacy staff - in conjunction with the Patient Care Area Monthly inspection form, review all crash carts for completeness in assigned areas. Document this review on the Patient Care Area Monthly Inspection form." On 5/5/11 at 8:00 a.m., Pharm 1 said emergency carts containing medications were checked monthly. The facility's policy was not consistent with State regulations. CCR, Title 22, 70263(f)(3) requires that the emergency medication supply "shall be inspected by a pharmacist at periodic intervals specified in written policies. Such inspections shall occur no less frequently than every 30 days."

On 5/9/11 at 2:00 p.m., the portable emergency container in the PACU for treating patients with malignant hyperthermia (MH) was inspected with Pharm 4 and RN 17. The list on the outside cover indicated that it should have contained five-50 ml vials (containers) of 8.4% sodium bicarbonate (to lower blood acidity) solution. It contained only 4 vials. At 2:50 p.m. the MH container (bag) in the Operating Rooms area contained four-50 ml vials of sodium bicarbonate 8.4% and 21 containers of dantrolene (muscle relaxant) 20 mg (milligrams). The list on the outside cover indicated that there were five vials of sodium bicarbonate and 20 containers of dantrolene.

4. On 5/2/11 at 2:40 p.m., the anesthesia code box in the Post Anesthesia Recovery (PACU) area was examined with RN (registered nurse) 1 and RN 2. It contained naloxone, 2 mg (milligrams)/2 ml with an expiration date of 4/11 and rocuronium, 10 mg/ml, 5 ml, with an expiration date of 6/7 (no year).

On 5/2/11 at 3:00 p.m., the refrigerator at Station 3 of Davis 12 Unit had a syringe of Epogen (an injectable medication for anemia) with an expiration date of 4/30/11. The Davis 12 Supply Room had five-100 milliliter Baxter mini-bags of 5% dextrose (sugar) in water (D5W) for intravenous administration. These bags were removed from the outer wrap and the opening dates were not listed on the containers (expiration date undeterminable). At 3:10 p.m., Pharm 5 said the Baxter mini-bags were good for 30 days once removed from the outer wrap.

On 5/2/11 at 3:20 p.m., the anesthesia cart in Operating Room 28 was examined with RN 2. It contained an opened (partially used) container of esmolol (for irregular heart rhythms), 100 mg/10 ml. It was labeled as a single dose vial (container) which should not be re-used after opening. An ophthalmology cart in the room contained a vial of epinephrine, 1 mg/ml, with an expiration date of 12/10.

On 5/2/11 at 3:30 p.m., the Supply Room of Tower 4 Nursing Unit had four unwrapped and undated 100-milliliter D5W mini-bags. Since there were no opening dates, the expiration dates were undeterminable.

On 5/2/11 at 3:45 p.m., intravenous (IV) solutions including 1000 ml lactated ringers (salt) solutions were observed with RN 2 in a warmer in the Operating Rooms area. RN 15 said the IV solutions were stable in the warmer and "ok" for 14 days. Several of the containers of lactated ringers solution were dated 5/22 which was 20 days in the future and more than 14 days. RN 15 said there were no temperature logs for the warmer and she said she did not have any information from the manufacturer regarding allowable temperatures and length of time their solutions could remain in the warmers without expiring (possibly deteriorated). RN 15 was advised to obtain this information from Baxter, the manufacturer of the lactated ringers solution. On 5/9/11 RN 16 provided the manufacturer's information which read "Solutions of volumes 150 ml or greater can be warmed in their plastic overpouches to temperatures not exceeding 40 degrees centigrade (104 degrees Fahrenheit), and for a period of no longer than 14 days." On 5/9/11 at 2:50 p.m., RN 16 said the hospital did not have a policy and procedures regarding the length of time and temperatures allowed for warming IV solutions.

On 5/2/11 at 4:05 p.m., mobile carts in the the Anesthesia Work Room were examined for expired drugs with RN 2. Pediatric cart 1 contained the following expired medications:

labetalol (decreases blood pressure), 20 mg/4 ml, expiration (exp.) date: 3/22/1
naloxone, 0.4 mg/ml, exp. date: 3/1/11
lidocaine, 5 mg/ml, 50 ml, opened but not dated
ketorolac (for pain), 30 mg/ml, opened but not dated
0.9 % saline (salt) solution, 50 ml, out of overwrap, not dated

Pediatric cart 2 in the Anesthesia Work Room contained the following expired medications:

lidocaine, 5 mg/ml, 50 ml, opened but not dated
succinylcholine (causes muscle paralysis), 20 mg/ml, 20 ml, one each with exp. dates: 4/22/11 and 5/2/11

On 5/3/11 at 9:20 a.m., expired medications were observed in Interventional Radiology with RN 2 and RN 14. The medications in room 1623 included epinephrine, 1 mg/10 ml, expiration date of 5/1/11 and sterile water for irrigation, 1000 ml, in the refrigerator with an expiration date of 2/10.

On 5/3/11 at 10:05 a.m., a 250 ml container of 5% dextrose solution with an expiration date of 3/11 was observed with RN 2 in the Computerized Tomography area.

On 5/3/11 at 10:20 a.m., 11 containers of barium sulfate (x-ray contrast solution), 454 grams each, were observed with RN 2 with expiration dates of 7/10.

On 5/4/11 at 2:00 p.m., vaccines including Adacel (tetanus, diptheria, pertussis) and influenza were observed in refrigerators in the Family Practice, Pediatrics, and Orthopedic clinics in the Ellison Building. On 5/5/11 at 11:00 a.m., Pharm 1 said clinic refrigerators were not on their central electronic monitoring system. He said there was a device which measured the highest and lowest temperatures in the refrigerators in the clinics. However, a review of the temperature logs in these clinics did not show temperatures being recorded on weekends when the clinics were closed. Hospital policy and procedure for ("Refrigerator/Freezer Requirements of Medications, Patient Nutrition and Laboratory Specimens and controls, Policy IV-57") included directions for temperature monitoring for areas not open 24 hours/day. The policy instructed: "When the unit has been closed, the minimum and maximum temperature is to be read and recorded manually the next business day". On 5/9/11 at 4:05 Pharm 1 said pharmacy personnel were not monitoring refrigerator temperatures in the clinics on weekends. Storage temperatures for medications requiring refrigeration potentially contribute to deterioration including decreased potency.

The hospital's policy and procedure ("Medications/Solutions/Vaccines in Single and Multiple-Dose Vials and Bottles, revised 12/14/10"), indicated: "Those products with special dating requirements shall be dated upon opening or mixing and discarded as recommended by the manufacturer, and no later than 28 days after opening."

Based on observations made in patient care areas, reviews of patient electronic medical records (EMRs), review of administrative documents, and staff interviews, the hospital failed to ensure patient safety when drugs and biologicals were not controlled and distributed in accordance with Federal and State regulations and standards of practice. The hospital failed
to develop and implement policies and procedures for safe and effective systems for distribution, dispensing and use of medications. These deficiencies occurred when the hospital failed to:

1. Establish a policy regarding droperidol that ensured patients with prolonged QTc intervals would not receive droperidol and that all patients would have a 12-lead ECG to assess the QTc interval and be monitored for two to three hours after receiving a dose of droperidol. Six out of twelve patients (Patients 3, 6, 7, 9, 10, and 14) received droperidol when records of 12-lead electrocardiograms (taken from one days to up to four months prior to the administration of droperidol) were available for review by the prescribing physician in the electronic medical record (EMR) that indicated that these patients had a prolonged QTc interval (the QT interval is an interval on the ECG and the QTc is a " corrected " interval) which is a listed contraindication to the use of this medication in the product labeling. Three out of twelve patients (Patients 5, 8, and 12) did not have a 12 lead electrocardiogram (ECG: a recording of the electrical voltage impulses of the heart recorded at various surfaces of the body) prior to administration of droperidol as recommended in the boxed warning. Two out of twelve patients (Patients 11 and 12) were not monitored via an ECG during and for at least 2 hours after the administration of droperidol following as recommended in the boxed warning.

Administration of the drug to patients with a known or suspected prolonged QTc interval is listed as a contraindication to use of the drug in the boxed warning section of the product labeling. Failure to use a 12-lead ECG to detect the presence of a prolonged QTc interval, use of the drug in a contraindicated fashion (when the physician would have reason to suspect that the QTc interval may be prolonged based on the history available to him or her in the EMR), and failure to monitor patients after receiving droperidol as recommended in the FDA approved boxed warning exposed these patients to the risk of developing torsades de pointes and/or exposed them to the risk of developing this arrhythmia in an unmonitored situation. The use of droperidol contrary to the boxed warning recommendations and contrary to the then current hospital guidelines was an issue that resulted in deficient findings and the calling of a state immediate jeopardy under the hospital pharmaceutical services during a licensing survey conducted at the hospital during October 2007.

2. Ensure pharmacists reviewed all new medication orders before nurses removed the drugs from ADCs (automated drug dispensing cabinets) and administered them to patients. New medication orders were reviewed by pharmacists in order to increase patient safety and confirm the correct choice of medication. This occurred when ADCs were not "profiled" (defined below) and when unnecessary ADC overrides (defined below) may have occurred. Overrides occurred with many dangerous drugs.

3. Determine the disposition of medications when the hospital did not proceed to follow-up on all discrepancies (defined below) in medication counts in their ADCs. This absence of accountability for many dangerous drugs in the ADCs may have resulted in medication errors and diminished staff safety when doses of medications were in excess or missing.

4. Have pharmacists conduct unit inspections according to California Code of Regulations (CCR) requirements. Unit inspections by pharmacists can prevent the hospital from unsafe practices such as administering expired medications and medications which have been stored at incorrect temperatures or are otherwise deteriorated.

5. Prepare and maintain portable emergency medication supplies according to CCR requirements. Failures included: not conducting inspections by pharmacists; lack of accurate and clearly visible lists of medications and the expiration date of the earliest drug to expire on the outside covers of portable emergency containers; not ensuring that all of the medications on the lists were present in the portable containers; and not having a pharmacist seal the emergency medications in the containers. These failures potentially increased the probability that patients would not survive a medical emergency when such medications were needed.

6. Ensure that staff were sufficiently knowledgeable and trained to treat malignant hyperthermia (MH), a potentially fatal medical emergency requiring rapid and effective treatment. The hospital also failed to inform hospital caregivers of the locations of these potential life-saving medications used to treat MH.

7. Ensure that nurses accurately administered intravenous (IV) insulin to two of two reviewed patients whose IV insulin records demonstrated dosing errors. Administering incorrect doses of insulin potentially increased the probability a patient would suffer from hyperglycemia (high blood sugar) and hypoglycemia (low blood sugar).

8. Ensure that Emergency Department nurses were adequately trained to draw into syringes correct doses of emergency medications for pediatric patients. This occurred when two of three nurses who were tested made errors. Medication errors during medical emergencies potentially placed patients at increased risk to suffer or fail to recover as a result of excessive or inadequate doses of emergency medications.

9. Ensure that directions for nurses administering potent IV medications were sufficiently detailed in order to prevent them from having to determine how to adjust doses for patients. Instructions for two medications, nitroprusside and nitroglycerin (lower blood pressure), were reviewed. These instructions for nurses provided wide parameters for nurses to determine dosage changes. The parameters did not include dosage adjustment instructions based on patient blood pressure readings or place a lower limit on a patient's blood pressure. These failures increased the unwanted opportunity a patient would suffer from excessive or inadequate doses of medications.

Findings:

1. Droperidol (brand name Inapsine) is a medication with a Food and Drug Administration (FDA: the institution that regulates the approval, labeling, and all regulatory issues regarding medications in the United States) approved use to treat nausea and vomiting. This medication has a boxed warning (also called a "black box" warning) which is the most serious warning required by the Food and Drug Administration to be placed in the product labeling (also called the package insert) for a medication. Boxed warnings document potential problems that can lead to serious injury or death. The boxed warning documents that use of droperidol has been associated with prolongation of the QT interval and/or torsades de pointes (an uncommon but potentially fatal irregular heartbeat).

The QT interval is seen on the ECG of the heart and is the period the lower chambers of the heart undergo electrical stimulation and then "recharge" so to speak. If this period is extended too long, abnormal heart beats may occur including torsades de pointes which can lead to more serious arrhythmias. The boxed warning makes the following points:

i. Cases of QT interval prolongation and/or torsades de points can occur at recommended or below recommended doses.

ii. This has occurred in patients without known risk factors for QT interval prolongation and some cases have been fatal.

iii. All patients should be screened using a 12-lead ECG to check for pre-existing QT prolongation and if present, droperidol should NOT be used (emphasis is in the boxed warning and defined as a QTc more than 440 milliseconds in males or 450 milliseconds in females). The QTc is a "corrected" QT interval as the QT interval can vary with heart rate.

iv. Patients who receive the medication should be monitored via an ECG (i.e. telemetry) prior to and for 2 to 3 hours after receiving the drug.

A review of 12 closed patient electronic medical records revealed the following issues in 10 of those records:

a. On 5/4/11 a review of Patient 3's EMR indicated Patient 3 was a female and her EMR contained a "Cardiac Catheterization Report" dated 3/21/11. The report documented that the patient received 1.25 milligrams (mg) of droperidol intravenously (IV) at 2:44 p.m. on that date. On 5/4/11 at 10:37 a.m. during an interview of RN 19 at that time, she stated that she had worked in the Cardiac Cath Lab for 19 years. She stated that a physician would give a verbal order for the droperidol and the staff nurse would administer it to the patient. She stated in her experience droperidol was commonly used. She stated that all patients in the Cath Lab got a 12 lead ECG before catheterization and staff would use what was available in the EMR: they would not repeat the 12 lead ECG if one was available in the EMR. Patient 3's EMR contained three 12 lead ECGs all done on 3/17/11 at 12:18 p.m. 11:30 p.m. and 11:31 p.m. The QTc intervals reported out on these ECGs were: 402 milliseconds (msec), 491 msec, and 484 msec respectively. Thus the last two ECGs done closest in proximity to the dosing of the droperidol on 3/21/11 were above the threshold for QTc prolongation for females (450 ms) as documented in the boxed warning at which level Patient 3 should not have received this medication according to the boxed warning.

b. On 5/4/11 a review of Patient 6's EMR indicated Patient 6 was a male and his EMR contained a "Cardiac Catheterization Report" dated 4/22/11. The report documented that Patient 6 received 1.25 mg of droperidol IV at 7:56 a.m. on that date. Patient 6's EMR contained two 12 lead ECGs done on 1/27/11 at 1:49 p.m. and 3/17/11 at 3:42 p.m. The QTc intervals reported out on these ECGs were: 461 msec and 459 msec respectively which was above the threshold for QTc prolongation for males (440 msec) as documented in the boxed warning at which level Patient 6 should not have received this medication according the boxed warning. During an interview of RN 20, she stated that these ECGs would have been available for reading by the physician as they had "scan" dates (dates scanned into the EMR by staff) of 2/16/11 and 3/29/11 respectively.

c. On 5/4/11 a review of Patient 7's EMR indicated that Patient 7 was a male and received 1.25 mg of droperidol IV at 5:30 a.m. on 2/5/11. This occurred in the Emergency Department (ED). The patient had been placed on a 3 lead ECG monitor at 2:54 a.m. There was no record of a 12 lead ECG done during the ED visit on 2/5/11 but Patient 7's EMR contained a 12 lead ECG done on 1/7/11 at 9:22 p.m. on which the recorded QTc was 444 msec. This was above the threshold for QTc prolongation for males (440 msec) as documented in the boxed warning at which level Patient 6 should not have received this medication. During an interview of RN 20 at that time she stated that staff had scanned the 1/7/11 ECG into the EMR on 1/8/11 so that it would have been available for the ED physician to review.

d. On 5/5/11 a review of Patient 9's EMR indicated that Patient 9 was a male and received 1.25 mg of droperidol IV on 3/6/11 at 10:30 a.m. and again on 3/7/11 at midnight and 9:23 p.m. while he was a patient in the Cardiac Care Unit (a unit for intensive care of patients with telemetry monitoring and one to one nursing care). Patient 9's EMR contained a 12 lead ECG done on 3/5/11 at 8:55 a.m. on which the recorded QTc interval was 487 msec. This was above the threshold for QTc prolongation for males (440 msec) as documented in the boxed warning at which level Patient 9 should not have received this medication. Staff had scanned this ECG into Patient 9's EMR on 3/5/11 at 9:40 p.m. so it would have been available for review by a physician on 3/6/11 before he received his droperidol.

e. On 5/5/11 a review of Patient 10's EMR indicated that Patient 10 was a female and received 1.25 mg of droperidol IV on 3/11/11 at 3:03 p.m. and again on 3/12/11 at 7:38 p.m. while she was a patient in the Surgical Intensive Care Unit (a unit for intensive care of patients with telemetry monitoring and one to one nursing care). Patient 10's EMR contained a 12 lead ECG done on 3/9/11 at 4:35 p.m. on which the recorded QTc interval was 470 msec. This was above the threshold for QTc prolongation for females (450 msec) as documented in the boxed warning at which level Patient 10 should not have received this medication. Staff had scanned this ECG into Patient 9's EMR on 3/10/11 at 3:14 p.m. so it would have been available for review by a physician on 3/11/11 before Patient 10 received her droperidol.

f. On 5/9/11 a review of Patient 14's EMR indicated that Patient 14 was a female and received 2.5 mg of droperidol IV on 4/2/11 at 9:44 a.m. while she was a patient in the ED. Patient 14's EMR contained three 12 lead ECG done on 3/6/11 at 10:04 a.m., on 3/28/11 at 7:44 a.m. and on 3/31/11 at 5 a.m. on which the recorded QTc intervals were 454 msec, 469 msec, and 454 msec respectively. This was above the threshold for QTc prolongation for females (450 msec) as documented in the boxed warning at which level Patient 14 should not have received this medication. During an interview of RN 20 on this date at 3:45 p.m. she stated that all of these ECGs would have been available for review by the physician before the droperidol was given.

g. On 5/4/11 a review of Patient 5's EMR indicated that Patient 5 received 1.25 mg of droperidol IV at 2:44 p.m. on 3/31/11. The patient had undergone a study of the vasculature in one of her legs even though the study was done in the Cardiac Cath Lab. There was no evidence in the EMR that a physician had ordered and staff had done a 12 lead ECG.

h. On 5/5/11 a review of Patient 8's EMR indicated that Patient 8 received 1.25 mg of droperidol IV on 2/17/11 at 6:10 p.m. while he was recovered in the Post Anesthesia Care Unit (PACU: unit where patients are watched after surgery to make sure they are stable before transfer to a regular hospital floor). There was no record of a 12 lead ECG in the EMR and this was verified by interview at 10:09 a.m. on this date with RN 20.

i. On 5/9/11 a review of Patient 12's EMR indicated that Patient 12 received 2.5 mg of droperidol IV on 2/7/11 at 2:55 a.m. and 2.5 mg IV at 8:05 a.m. on the same date while she was a patient in the ED. The EMR contained no record of a 12 lead ECG done before Patient 12 received her droperidol. Also, there was no evidence (confirmed on 5/10/11 at 12:31 p.m. with RN 20) of any ECG monitoring after Patient 12 received her droperidol.

j. On 5/9/11 a review of Patient 11's EMR indicated that Patient 11 received 2.5 mg of droperidol IV on 2/25/11 at 5:22 p.m. while she was a patient in the ED. Patient 11's EMR contained 12 lead ECGs done on 2/4/11 at 11:59 a.m. and 2/25/11 at 5:11 p.m. on which the recorded QTc intervals were 417 msec and 436 msec respectively which were NOT prolonged according to the boxed warning for droperidol. However, there was no evidence (confirmed on 5/10/11 at 2:21 p.m. with RN 20) of any ECG monitoring after Patient 12 received her droperidol.

On 5/4/11 a review of Policy and Procedure entitled PATIENT CARE STANDARDS IV-55 (Revised 1/4/11) indicated that it had a "Section 1" entitled "Medication Administration". Under "Procedure P. 2" the policy provided the following information and procedures regarding IV (intravenous) droperidol:

"a. The FDA has issued a black box warning about droperidol specific to prolonged QTc..."

"b. Measure and document the QTc interval prior to administration for patients with known QTc prolongation or for single doses greater than 0.1 mg/kg or 2.5 mg. If greater than 440 msec for males or 450 msec for females or 25% greater than pre-droperidol QTc, do NOT (capital letters in policy) administer and notify the prescriber. The discussion with the prescriber and any follow-up action should be documented in the Notes section of the EMR."

The only way to measure a QT interval (and thus calculate a QTc interval) is to do an ECG. The policy made no provision to ensure that an ECG be done to determine if the QTc interval was prolonged in any given patient if there was no previous record of an ECG in a given patient's clinical record. As written the policy would not prevent the administration of droperidol to a patient with a prolonged QTc interval if, after being informed by staff of the prolonged QTc interval, a physician elected to proceed with the administration of droperidol in such a patient. If it was administered in this case, all that the staff would be required to do would be to document "the discussion with the prescriber and any follow-up action in the Notes section ..." Finally, as written, it would be necessary for staff to determine if the QTc was prolonged based on the patient's gender in order to decide whether or not to give the dose.

On 5/5/11 two nurses, who were interviewed in the ED, gave the following responses:

a. At 8:38 a.m. during an interview of RN 22, she stated that a QTc greater than 380 msec in males and greater than 410 msec in females would be a reason not to give droperidol.

b. At 8:38 a.m. during an interview of RN 23, she stated that a QTc greater than 450 msec in males and 440 msec in females would be a reason not to give droperidol.

The two nurses did not correctly identify the boxed warning QTc thresholds at which droperidol should not be given (the third one did so). However of those two, the one answer was low for both males and females so a patient with a prolonged QTc would not have received droperidol before a physician was contacted. The second nurse had the gender values for the QTc reversed and hence male patients could potentially have received droperidol when they should not have.

On 5/10/11 at 1:22 a.m. during an interview of Pharm 1 he stated that in 2007 the hospital was not following its own protocol and that at that time it was not to be used first line for the treatment of nausea and vomiting. With regard to the boxed warning, he said the drug had the potential to cause QT prolongation. He said that in general he thought that the boxed warning was "excessive". He stated that when he said "I think" he was speaking from the perspective of the Pharmacy and Therapeutics (P&T) Committee (the committee with oversight of the pharmaceutical services in the hospital and which develops policies and procedures establishing safe and effective systems for the procurement, storage, distribution, dispensing and use of drugs in the hospital) and the literature that was available on the subject. He stated that the P&P for droperidol provided on 5/4/11 had been modified "last week" by adding the words "risk factors for" in the sentence "Measure and document the QTc interval prior to administration for patients with known risk factors for QTc prolongation ..."

Pharm 1 produced a copy of the P&T Committee meeting minutes from 12/19/07. A review of the minutes indicated it documented that droperidol may be used first line and administered in doses less than or equal to 2.5 mg with no additional monitoring in patients with no known risk factors for QT prolongation. The prescriber decides what the risk factors are. Pharm 1 stated that these were the current hospital "guidelines" and that the "no known risk factors" part of the minutes is what was used to modify the policy and procedure.

Pharm 1 produced a copy of the P&T Committee meeting minutes from 11/28/07. A review of the minutes indicated it documented that the committee felt that the black box warning with the use of droperidol was "extreme" ..." The committee recommended no ECG monitoring with doses less than or equal to 2.5 mg.

Pharm 1 stated that it was more typical for the hospital to develop medication guidelines and incorporate them into order sets. He said that a guideline was not a policy, it was a recommendation but if specific in nature, they would be expected to be followed. He pointed out that the 1/4/11 PATIENT CARE STANDARDS IV-55 was not congruent with the above droperidol guidelines. He stated the policy was developed after the "guidelines" and the two should have been consistent but they were not. He could not say during the interview which of these, a guideline or a policy and procedure was the determinate factor for action. He didn't know.

Pharm 1 stated he felt that an increased QTc interval was a risk factor with the use of droperidol and therefore would be a reason to contact the prescriber per the PATIENT CARE STANDARDS IV-55.


23014


2. Automated drug dispensing cabinets:
a. On 5/2/11 at 10:00 a.m., Pharm (pharmacist) 1 and Pharm 2 said hospital automated drug dispensing cabinets (ADCs) were not "profiled" in medical procedural areas such as the gastrointestinal laboratory and the cardiac catheterization laboratory, Labor and Delivery, Post Anesthesia Recovery (PACU), and the Emergency Department. The ADCs were electronically connected to the hospital computer system. Profiled ADCs were those which required a pharmacist to review prescribers' new medication orders and approve them electronically before a nurse could remove the ordered medications from the ADC. Prescribers (for example, physicians) entered new medication orders directly into the hospital computer and pharmacists were able to view them on the computer screen. Medications obtained from ADCs which were not profiled did not require a pharmacist to review the new medication orders before removal. Pharmacists reviewed new medication orders for several safety and efficacy (effective drug for the disease) reasons which included: ensuring the correct medication and dose was ordered and to avoid unwanted drug interactions, adverse drug effects (side effects), drug allergies, and medication errors. Medications removed for patients from ADCs which were not profiled did not benefit from these safety and efficacy checks.

Examination of the hospital's medication error reduction plan ("UC Davis Medical Center Medication-Related Error Reduction Plan December 19, 2001") indicated on page 9 that, "Nurses do not remove medication from PYXIS (ADC) until a pharmacist has reviewed the order." Under"Comment" on this listing was the conclusion, "Done, exception is emergency situations."

b. On 5/2/11 at 10:30 a.m., Pharm 1 said the "override" function was used for about one percent of the medications removed from ADCs. Overrides occurred on profiled ADCs when medications were removed before a pharmacist had reviewed the new medication orders. Overrides prevented pharmacists from reviewing new medication orders for safety and efficacy checks before patients received the medications.

A review of the hospital's report on overrides for January, 2011 ("Rx Auditor, Override Summary Analysis, January 2011") demonstrated that the number of overrides were 1,980 which were 1.05 percent of 188,492 "dispense events". A dispense event occurred each time a nurse removed medication from an ADC, usually for a particular patient. The February, 2011 report listed 2,036 overrides and the March 2011 report listed 1,708 overrides.

The hospital provided a list of 93 medications which could be removed on override from all ADCs and a separate list of 53 medications which could be removed on override only from designated ADCs. Examples of medications on these lists included muscle paralyzing agents (succinylcholine, vercuronium), narcotic type analgesics (fentanyl, hydromorphone, morphine), sedatives (lorazepam, diazepam, phenobarbital), medications to increase blood pressure, decrease blood pressure, and medications to increase the heart rate. Many of the medications on the lists could have been lethal (could kill you) if they were administered in overdose or to the wrong patient.

Hospital policy and procedures ("Automated Medication Dispensing Machines, revised 5/13/09") read, "these [override] medications are approved based on consideration of urgency of need, time to onset of action, potential dangers of administration and lack of need for dosage adjustment in renal/hepatic dysfunction." Some of the reasons listed in the computer for medications removed on override ("Override Detail Response Analysis, March, 2011") included: "Procedural use; RN (registered nurse)/RT (respiratory therapist) clinical evaluation of patient status; to relieve pain; MD states one dose needed urgently; no allergy to this medicine; to flush an IV (intravenous) catheter". Many of the overrides did not have a reason for removal and others were nonsensensical (example:"no-no action").

The American Society of Health-Systems Pharmacists (ASHP) has established the standards of practice for ADCs ("ASHP Guidelines on the Safe Use of Automated Medication Storage and Distribution Devices: Am J Health-Syst Pharm. 1998; 55:1403-07"). The ASHP guidelines indicate that the automated system or device should ..."Provide for prospective, timely review of medication orders by a pharmacist at all appropriate decision points in the medication-use process, especially before administration of the first dose; and provide for the independent interpretation of the medication order by a pharmacist and a nurse." ASHP also points out that "Clearly stated organizational policies should be developed that limit access to medications before orders have been reviewed and approved by a pharmacist. Access to medications should be limited to the following cases: 1. The order has been reviewed and approved by a pharmacist. 2. The drug product has been approved by a multidisciplinary committee ... who agree that it has minimal risk for misadventures. 3. There is a clinically urgent need for the medication that outweighs the potential risk. .." Also, "Provision should be made for the retrospective review and reconciliation by a pharmacist of orders that were initiated without a pharmacist's review and approval." At 8:00 a.m. on 5/3/11, Pharm 1 said the hospital did not follow-up on overrides to determine whether the medications were removed according to hospital policy. He did, however, say that pharmacists retrospectively (after the override) followed-up on overrides to ensure that a prescriber's order had been entered into the computer.

3. On 5/2/11 at 10:30 a.m., Pharm 1 said the hospital did not follow-up on all "discrepancies" from ADCs. He said the hospital only followed-up to determine the disposition (what happened to the medication) of controlled substances discrepancies. Controlled substances included narcotic analgesics (for example, morphine) and sedative type medications (for example, phenobarbital and Valium). A discrepancy occurred when the number of doses of a medication in the ADC was discovered to be different from the number of doses listed in the computer attached to the ADC. The numbers of doses placed in the ADC were listed in the attached computer when the medication drawer or pocket in the ADC was filled. For example, discrepancies could occur when a nurse indicated in the computer that she removed one dose but actually removed two or more. Also, a discrepancy could occur if a nurse listed in the computer that she removed one dose but changed her/his mind and did not remove it. Potential problems attributable to discrepancies included drug diversion (employee taking medication for own use), administration of too many or too few doses, and other medication errors such as giving the medication to the wrong patient.

A review of the hospital's reports on ADC discrepancies from 2/1/11 through 4/28/11 showed that there were approximately 400 controlled substances discrepancies and approximately 530 discrepancies for other potentially dangerous medications. The controlled substances discrepancies listed the reasons for the discrepancies and the resolution. The non-controlled substances discrepancies did not list reasons for discrepancies nor resolutions.

The hospital's policy and procedures for ADCs ("Automated Medication Dispensing Machines, revised 5/13/09") did not address the issue of dealing with or resolving discrepancies for medications which were not controlled substances. Examples of these discrepancies included dangerous medications such as succinylcholine and rocuronium (cause muscle paralysis), atropine (increases the heart rate), propofol and dexmedetomidine (sedatives), nitroprusside (lowers blood pressure), nicotine patches, quetiapine (antipsychotic for mental illness), antibiotics, and heparin and warfarin (decrease blood clotting).

ASHP ("ASHP Guidelines on the Safe Use of Automated Medication Storage and Distribution Devices: Am J Health-Syst Pharm. 1998; 55:1403-07") indicates that "The organization should have a written plan for the monitoring and surveillance of medications accessed through automated systems. ... The plan should include identification of the data to be captured and the reports generated that are used to monitor medication use (data and reports may vary by drug categories and requirements for control and accountability). ... A description of the process for investigating trends in discrepancies and assigning responsibility for this."

4. At 10:20 a.m. on 5/2/11, Pharm 1 said Pharm Tech (pharmacy technician) 1 conducted all of the hospital nursing unit inspections with a pharmacist. The California Code of Regulations( CCR) Title 22, 70263(q)(10) read: "Drugs maintained on the nursing unit shall be inspected at least monthly by a pharmacist." There is no provision for anyone other than a pharmacist to conduct the unit inspections. Unit inspections by pharmacists can prevent the hospital from unsafe practices such as administering expired medications and medications which have been stored at incorrect temperatures or are otherwise deteriorated.

5. On 5/2/11 at 10:45 a.m., transport kits were observed in the pharmacy. Pharm 2 said the portable transport kits which contained medications were filled in the pharmacy. He said they were emergency kits but did not contain lists of medications on the outside cover of the container. He said the lists were on the inside of the container. CCR Title 22, 70263 (f)(2) required that "The emergency drug supply shall be stored in a clearly marked portable container which is sealed by the pharmacist in such a manner that a seal must be broken to gain access to the drugs. The contents of the container shall be listed on the outside cover and shall include the earliest expiration date of any drugs within."

On 5/2/11 at 1:30 p.m., RN 9 in the Medical Intensive Care Unit (MICU) said she did not know there was a list of medications on the outside cover of their portable emergency cart.

On 5/2/11 at 2:35 p.m., RN 1 in the Post Anesthesia Recovery Unit (PACU) area said the pediatric and adult airway boxes located there were portable emergency containers with medications inside. The adult box was opened and the following were examples of medications which were observed inside: lidocaine oral solution (for oral pain), vercuronium and succinylcholine (for muscle paralysis), and atropine (increases heart rate). There were 17 medications inside. The container did not have a list of medications and did not have an expiration date of the first medication to expire on the outside cover. MD (physician) 1 said she stocked (placed medications inside) the adult airway box and sealed it with a yellow plastic lock from a supply which was located in the PACU.

On 5/2/11 at 2:40 p.m., an anesthesia code box, which was located in the PACU, was inspected with RN 1 and RN 2. It was a portable emergency container with 24 medications inside. It was sealed with a yellow plastic lock and MD 1 said she placed medications inside the box and sealed the container. MD 1 said it was the hospital's practice for the anesthesiology resident to prepare the anesthesia code box with medications and equipment. The box did not contain a list of medications and the expiration date of the first drug to expire on the outside cover. It also contained the expired medications, naloxone (to reverse narcotic analgesics such as morphine) and rocuronium (for muscle paralysis).

On 5/3/11 at 10:50 a.m., an adult airway emergency box in the pediatric

Based on observations, staff interview, and administrative record review, the hospital failed to remove expired and other unusable medications from patient care areas. Administration of such medications could have produced unexpected (for example, reduced) effects when administered to patients. Unexpected effects of medications could have impaired and/or delayed a patient's recovery.

Findings:

On 5/2/11 at 2:40 p.m., the anesthesia code box in the Post Anesthesia Recovery (PACU) area was examined with RN (registered nurse) 1 and RN 2. It contained naloxone (to reverse narcotic-type analgesics), 2 mg (milligrams)/2 ml (milliliters) with an expiration date of 4/11 and rocuronium, 10 mg/ml, 5 ml, with an expiration date of 6/7 (no year).

On 5/2/11 at 3:20 p.m., the anesthesia cart in Operating Room 28 was examined with RN 2. It contained an opened (partially used) container of esmolol (for irregular heart rhythms), 100 mg/10 ml. It was labeled as a single dose vial (container) which should not be re-used after opening. An ophthalmology cart in the room contained a vial of epinephrine (increases blood pressure and heart rate), 1 mg/ml, with an expiration date of 12/10.

On 5/2/11 at 3:45 p.m., intravenous (IV) solutions including 1000 ml lactated ringers (salt) solutions were observed with RN 2 in a warmer in the Operating Rooms area. RN 15 said the IV solutions were stable in the warmer and "ok" for 14 days. Several of the containers of lactated ringers solution were dated 5/22 which was 20 days in the future and more than 14 days. RN 15 said there were no temperature logs for the warmer and she said she did not have any information from the manufacturer regarding allowable temperatures and length of time their solutions could remain in the warmers without expiring (possibly deteriorated). RN 15 was advised to obtain this information from Baxter, the manufacturer of the lactated ringers solution. On 5/9/11 RN 16 provided the manufacturer's information which read "Solutions of volumes 150 ml or greater can be warmed in their plastic overpouches to temperatures not exceeding 40 degrees centigrade (104 degrees Fahrenheit), and for a period of no longer than 14 days." On 5/9/11 at 2:50 p.m., RN 16 said the hospital did not have a policy and procedures regarding the length of time and temperatures allowed for warming IV solutions.

On 5/2/11 at 4:05 p.m., mobile carts in the the Anesthesia Work Room were examined for expired drugs with RN 2. Pediatric cart 1 contained the following expired medications:

labetalol (decreases blood pressure), 20 mg/4 ml, expiration (exp.) date: 3/22/11
naloxone, 0.4 mg/ml, exp. date: 3/1/11
lidocaine (for irregular heart rhythms), 5 mg/ml, 50 ml, opened but not dated
ketorolac (for pain), 30 mg/ml, opened but not dated
0.9 % saline (salt) solution, 50 ml, out of overwrap, not dated

Pediatric cart 2 in the Anesthesia Work Room contained the following expired medications:

lidocaine, 5 mg/ml, 50 ml, opened but not dated
succinylcholine (causes muscle paralysis), 20 mg/ml, 20 ml, one each with exp. dates: 4/22/11 and 5/2/11

On 5/3/11 at 9:20 a.m., expired medications were observed in Interventional Radiology with RN 2 and RN 14. The medications in room 1623 included epinephrine, 1 mg/10 ml, expiration date of 5/1/11 and sterile water for irrigation, 1000 ml, in the refrigerator, expiration date of 2/10.

On 5/3/11 at 10:05 a.m., a 250 ml container of 5% dextrose (sugar) solution with an expiration date of 3/11 was observed with RN 2 in the Computerized Tomography area.

On 5/3/11 at 10:20 a.m., 11 containers of barium sulfate (x-ray contrast solution), 454 grams each, were observed with RN 2 with expiration dates of 7/10.

On 5/4/11 at 2:00 p.m., vaccines including Adacel (Tdap) and influenza were observed in refrigerators in the Family Practice, Pediatrics, and Orthopedic clinics in the Ellison Building. On 5/5/11 at 11:00 a.m., Pharm 1 said clinic refrigerators were not on their central electronic monitoring system. He said there was a device which measured the highest and lowest temperatures in the refrigerators in the clinics. However, a review of the temperature logs in these clinics did not show temperatures being recorded on weekends when the clinics were closed. Hospital policy and procedure for ("Refrigerator/Freezer Requirements of Medications, Patient Nutrition and Laboratory Specimens and controls, Policy IV-57") included specific instructions for monitoring temperatures in areas that are not open 24 hours/day. The policy included: "When the unit has been closed, the minimum and maximum temperature is to be read and recorded manually the next business day". On 5/9/11 at 4:05 Pharm 1 said pharmacy personnel were not monitoring refrigerator temperatures in the clinics on weekends.


27000


On 5/2/11 at 1:30 p.m., a tour of the Davis 12 Nursing Unit was conducted in the presence of RN 17 and RN 18. Two opened vials of tuberculin (PPD, product used to test for tuberculosis) were observed in the refrigerator at Station 1. One vial did not have an opening date; the other had an expiration date of 3/1/11. RN 18 said multi-dose (for multiple uses) vials had to be dated with opening date but she was unsure how long they were good for once opened.

On 5/2/11 at 3 p.m., the refrigerator at Station 3 of Davis 12 Unit had a syringe of Epogen (an injectable medication for anemia) with expiration date of 4/30/11.

During the same tour, the Davis 12 Supply Room had five-100 milliliter Baxter mini-bags of 5% dextrose (sugar solution) in water (D5W) for intravenous administration. These bags were removed from the outer wrap and were without the opening date (expiration date undeterminable). At 3:10 p.m., Pharm (pharmacist) 5 said the Baxter mini-bags were good for 30 days once removed from the outer wrap.

On 5/2/11 at 3:30 p.m., the Supply Room of Tower 4 Nursing Unit had four unwrapped and undated 100-milliliter D5W mini-bags. Since there were no opening dates, the expiration dates were undeterminable.

Based on observation, staff interviews, and document review, the hospital failed to maintain appropriate temperature controls in one refrigeration unit, the investigational drug refrigerator (REF 1) in the pharmacy, for 10 days in April 2011. The failure to main refrigeration unit may result in compromised medication integrity and stability, which could result in unsafe and/or ineffective medications for patients.

Findings:

On 5/2/11 at 10 a.m., during a tour to the Central Pharmacy, the Director of Pharmacy (Pharm 1) said the hospital had a electronically monitoring system (called the Metasys) for medication refrigerators and freezers in the hospital's pharmacies, monitored by the Plan Operations & Maintenance (POM). He explained that when the temperature went out of range, the system would set off an alarm. The POM operator would call the pharmacist or send a personnel to investigate and revolve the issue.

On 5/5/11, the Pharmacy Assistant Manager (Pharm 2) provided the temperature readings of the medication refrigerators and freezers in the hospital's pharmacies from 2/15/11 to 4/30/11.

On 5/5/11 at 11:30 a.m., Pharm 2 said the hospital set the temperature parameters for all medication refrigerators at 2?C to 8?C. Review of the Central Pharmacy's Refrigerator 1 (REF 1) with Pharm 2 revealed the temperatures were continuously out of range for 10 days from 4/11/11 to 4/21/11, ranging from 1.38?C to as low as -1.47?C (below freezing, which is 0?C). When asked if he was aware of the out-of-range temperature for this time period, Pharm 2 said, "I don't know."

On 5/5/11 at 12 p.m., a visit to the Central Pharmacy was conducted with Pharm 2 and a Staff Pharmacist (Pharm 6). Observation of REF 1 reflected it contained various investigational drugs, each in bulk quantities (the hospital had ongoing investigational studies). Pharm 6 said the temperature requirement for these investigational drugs was from 2 to 8?C.

During the same interview, Pharm 6 said the temperature for REF 1 went out of range on 4/20/11, after which he immediately notified POM. He explained, in the 4/20/11 e-mail communication with the POM Manager (POM MGR), he was told the temperature went out of range on 4/11/11 at 4:52 p.m. and the system did set off an alarm, however, the POM operator failed to dispatch the alarm by failing to send a personnel to investigate the affected site.

On 5/5/11 at 2 p.m., the POM MGR was interviewed in the presence of Pharm 2. He confirmed Pharm 6's statement. He said, per hospital's policy and procedure, it was the responsibility the POM to operate, monitor, and respond to alarms, of the hospital's automation system including pharmacy's refrigerators. He said "the operator failed to dispatch the alarm" on 4/11/11, and therefore the system "stayed in alarm". However, it would not stay active as a "pop-up" on the computer screen, but remained in the "stay alarm" functionality, which was not visible to the operator, therefore it was not acted upon until Pharm 6's notification. As a result, the affected refrigerator (REF 1) stayed out of range for 10 days, from 4/11/11 to 4/21/11.

The POM MGR provided the hospital policy and procedure entitled "METASYS SYSTEM OPERATION AND ALARMS", dated 9/06. It read, "Plant Operations & maintenance is responsible for the monitoring of equipment which is connected to the campus wide 'Metasys' Building Automation System (BAS).... It it the responsibility of Central Plant Operators to properly monitor and process alarm conditions when reported through the BAS."