ISMP Quarterly Action Agenda - April-June 2003

Click here for printer friendly version

One of the most important ways to prevent medication errors is to proactively seek and use knowledge from other organizations that already have experienced similar problems. To promote such a process, administrative staff and an interdisciplinary committee at each practice site should review the following material to prompt discussion and stimulate action to reduce the risk of errors. The following selected items appeared in the ISMP Medication Safety Alert! between April and June 2003. Each item includes a description of the problem, recommendations for safe medication practices, and the issue number (in parentheses) to locate additional information as desired. Many product-related problems can be visualized in the ISMP Medication Safety Alert! section of our web site. ACPE credit is available for pharmacists and nurses.

• How fast is too fast for IV push medications (10)

  Problem: A study in the UK suggested that the most common error with IV medications is too rapid administration.
  Recommendation: Provide maximum administration rates for medications that have a high risk of adverse effects when given too rapidly. Place these on pharmacy labels, computer screens, MARs, in PDAs and in chart form. An informative chart is being prepared for future publication in Hospital Pharmacy.

• Oral vancomycin does not treat systemic infections (10)

  Problem: We have received several reports of switching patients with systemic infections from IV to oral vancomycin at discharge. However, the oral form of vancomycin is not absorbed systemically.
  Recommendation: Computer screens and MARs should alert professional staff that oral vancomycin is not absorbed systemically.

• Dosing error with dextran (INFED, DEXFERRUM) (13)

  Problem: The Washington Manual of Medical Therapeutics, 30th edition, implies that the patient's actual body weight, not the lean body weight, should be used to calculate the dose. This led to an overdose for an obese patient.
  Recommendation: Design and implement a dosing protocol for this medication and have a pharmacist verify all dose calculations based on the patient's lean body weight before dispensing the product.

• Don't confuse lipid-based and conventional amphotericin B (13)

  Problem: Cases of confusion between lipid-based and conventional amphotericin B continue to be reported. In the latest report of confusion, a patient received the conventional product at a dose prescribed for the lipid-based product, AMBISOME (amphotericin B liposomal).
  Recommendation: Avoid storing conventional and lipid-based products in close proximity. Prescribers should refer to lipid-based products by their brand names. Place appropriate warnings about the potential for confusion in the pharmacy computer system. Require an independent double-check if multiple vials of amphotericin B are needed for preparation.

• Converting oral hydromorphone to the IV route (10)

  Problem: A physician converted an oral hydromorphone dose of 2-4 mg every 4 hours to the IV route at the same dose and frequency without recognizing that the proper oral to IV conversion ranges from 3:1 to 5:1. The patient experienced respiratory depression and required naloxone.
  Recommendation: Provide staff education about opiate dosing and update annually. Post oral to IV conversion charts for opiates on patient care units. Have pharmacists review all medication orders prior to administration.

• Digoxin dose limits in the computer (10)

  Problem: After adjusting the dose of digoxin for a patient with renal insufficiency, a pharmacist entered "0.63 mg" every 48 hours instead of "0.063 mg" and the patient received a ten-fold overdose as a result.
  Recommendation: Ensure that dose limits for digoxin are set in the pharmacy computer. Also, consider expressing digoxin doses in mcg.
  

• HIPAA and listing an indication on prescriptions (9)

  Problem: There's a common misperception that listing a medication's purpose on a prescription would violate the HIPAA Privacy Rule.
  Recommendation: According to the Department of Health and Human Services, supplying this very important information on the prescription does not violate the rule. Although this is "Protected Health Information," it is not a violation to communicate it to another health provider for treating the patient.

• LANTUS (insulin glargine [rDNA origin]) and LENTE (insulin zinc suspension) (10)

  Problem: In the past, it was easier to avoid confusion between Lantus and other insulin, such as Lente, because of its unique bedtime administration. Now FDA has approved a flexible daily dosing schedule, making it more difficult to differentiate Lantus from Lente, especially if orders are handwritten.
  Recommendation: Alert staff to the more flexible dosing regimen. Also, educate patients and staff not to mix Lantus with other insulins.

• The dilemma between evidence-based and obvious patient safety strategies (7)

  Problem: Some practitioners have asked if evidence exists to support the Joint Commission's National Patient Safety Goal related to prohibition of abbreviations and dose expressions that have led to medication errors.
  Recommendation: Scientific research does not exist to prove the effectiveness of this error reduction strategy, but evidence for its adoption is nonetheless obvious. Similar to how advances were made in aviation and nuclear power safety, patient safety requires the adoption of strategies that make sense and represent the obvious right thing to do.

• Drug protocols require groundwork (9)

  Problem: Problems arose after a physician obtained an insulin infusion protocol from another hospital and sought quick approval for its use. The new protocol called for dose adjustments every 2 hours, which resulted in labile blood sugars. It was later discovered that the original protocol was intended for patients who were receiving tube feedings or TPN.
  Recommendation: Allow adequate time to properly review a new protocol before its adoption. Perform a failure mode and effects analysis to determine vulnerable areas and develop measures to gauge its effectiveness on a small group of patients before spreading its use to all applicable patients.

• TNKase (tenecteplase) and ACTIVASE (alteplase) mix ups (11)

  Problem: Practitioners have confused Activase, which is commonly referred to as t-PA, with TNKase, which is sometimes called TNK or TNK-t-PA. Use of these abbreviations led to three overdoses. In each case, TNKase was given to a stroke patient at a dose appropriate for Activase, the intended drug.
  Recommendation: Avoid using abbreviations for these thrombolytics. Minimize the variety of thrombolytics on the formulary and standardize their use. Require an independent double-check before use if these drugs are not dispensed from the pharmacy (where they should also be double-checked).
  

• Fatal reports of intrathecal vincristine continue (9)

  Problem: Recently, the New Jersey media reported two fatalities involving intrathecal administration of vincristine instead of the prescribed medications, cytarabine in one case, and methotrexate in the other.
  Recommendation: Consider dispensing vincristine in an IV bag with proper warning labels and an attached IV administration set to differentiate it from intrathecal syringes. Administer intrathecal medications in a specified location and have pharmacy deliver them right before use. Never deliver intrathecal medications with IV medications. Require at least two health professionals to independently verify the drug and dose before administration.

• Ephedrine injection confused with epinephrine (8)

  Problem: Similar names, indications and appearances have led to mix-ups between epinephrine and ephedrine ampuls, resulting in serious adverse outcomes for two obstetrical patients.
  Recommendation: Differentiate these names on all communication media, including pharmacy and nursing unit storage bin labels, computer screens and MARs, using tall-man lettering (EPINEPHrine, ePHEDdrine). When possible, use prefilled syringes. If using automated dispensing cabinets, screens should caution users about this potential problem. Have pharmacy prepare all infusions and bolus doses of these drugs.

• METHERGINE (methylergonovine maleate) injection and BRETHINE (terbutaline sulfate) injection (8)

  Problem: These medications are packaged in amber clear plastic unitdose blisters that make the products look nearly identical and make reading the label difficult. Several mix-ups have been reported.
  Recommendation: FDA and the manufacturers are aware of the problem, but changes have not occurred despite numerous practitioner complaints spanning 3 years. We'll continue to push for packaging changes but until that occurs, if you must use these drugs, alert staff to the problem, store these products separately, and apply label reminders to caution against confusion.

• Tetanus-diphtheria toxoid and tuberculin purified protein derivative (9)

  Problem: Groups of patients were given Td toxoid intradermally instead of PPD, resulting in false positive PPD readings and unnecessary administration of isoniazid. Similar names and packaging may have contributed to the error.
  Recommendation: Purchase these products from different manufacturers when possible. Also, remove Td from unit stock if feasible.
  

• Outbreak of infections linked to multiple-dose lidocaine vials (12)

  Problem: A recent outbreak of Staphylococcus aureus infections has been associated with multiple-dose vials of lidocaine used during injections into joints. Investigators also found that refrigeration of the lidocaine vials after initial use was a contributing factor (refrigeration actually prolongs survival of any bacterial contaminants).
  Recommendation: Use prefilled lidocaine syringes or single-dose vials in areas outside the OR. If multiple-dose vials are used, do not refrigerate them.

• Pre-procedutre sedation by parents (10)

  Problem: A pharmacist misinterpreted a prescription for chloral hydrate liquid, "30 minutes before a child's office visit" as "30 cc before office visit." Subsequently, the mother of a 17-month-old gave an overdose to the child just before leaving for the physician's office. The child became comatose.
  Recommendation: Never administer pre-procedure sedation to a child at home. Skilled medical personnel and resuscitation equipment should always be present before sedation, and patients must be closely monitored.

• Blood pressure monitor tubing may connect to IV ports (12)

  Problem: The tubing leading from blood pressure monitors to certain disposable blood pressure cuffs may have a male Luer connection. If so, there is a serious risk that it may connect to a needleless Y site connection integrated with certain IV administration sets. If such a connection is made, the patient is at risk of air embolism once the monitor attempts BP cuff inflation. Several fatal cases have occurred.
  Recommendation: Ensure that disposable blood pressure cuffs do not have female Luer connectors. Place blood pressure monitors on a different arm than IV catheters, and remove IV catheters as soon as they are no longer needed.