Follow-up study of medication errors reported to the Vaccine Adverse Event Reporting System (VAERS).

Background: A study was done to determine if the apparent medication errors found in the Vaccine Adverse Event Reporting System (VAERS) database are true errors, and if true errors are found, to determine what corrective action was taken. Furthermore, if a true error did not occur, we wanted to determine at what point the misinformation was entered into the system.

Methods: The VAERS database was searched for reports received between July 1, 2001 and June 30, 2002 which had either been classified as "error" or the word "error" appeared in the text of the report. The database was also searched for reports which indicated that the measles-mumps-rubella (MMR), diphtheria-tetanus-pertussis (DTP), diphtheria-tetanus-acellular pertussis (DTaP) or diphtheria-tetanus-pertussis-haemophilus (DTPH) vaccinations had been administered at an age outside of the usual recommendation.

Results: A total of 119 reports of possible errors were found. Follow-up was successful in 102 (86%) cases. Additional information obtained showed that 26 cases were actual medication errors. Seventy-six cases were not actual medication errors; 9 cases were physician decisions, 37 cases were data entry errors and 30 cases were reporter errors.

Conclusion: The nature of the actual errors was similar to those reported previously; wrong inoculum, improper interval, wrong route of administration, and overdose. Many errors could have been prevented by more attention to detail. Remedial action usually consisted of retraining. The new requirement that all medications be bar-coded, purchasing products from different manufacturers and segregation of vials may help prevent vial confusion.

Key Words: vaccine, medication error, surveillance

**********

Since the publication of the Institute of Medicine (1) report on medication errors, there has been increased interest and awareness of this serious problem which is estimated to cause 100,000 hospital admissions per year in the United States. In order to reduce the number of medication errors it is vital to determine how, when and why errors occur.

The Vaccine Adverse Event Reporting System (VAERS) (2) was established to receive reports of possible vaccine adverse events postlicensure. It is a passive reporting system with all of the advantages and disadvantages typical of passive reporting, including under reporting, incomplete reports and inaccurate information. (3) VAERS was not conceived as a means of detecting medication errors and there is no requirement to report medication errors to the FDA at this time. Nevertheless, medication errors are occasionally detected as part of the vaccine adverse event report review. A published review of VAERS found 49 medication errors in reports received between January 1994 to January 2001. (4) The errors were categorized as wrong inoculum (30%), improper interval between immunizations (20%), wrong route of administration (18%), preparation error (4%), polypharmacy (2%), and other (2%). (4)

A recent report describes approximately 100 incidences of patients receiving tetanus toxoid-containing vaccines and other vaccines instead of tuberculosis skin tests. (5) On the other hand, at least 8 people have received IM administration of tuberculin instead of vaccine. (5)

This report is a follow-up study to determine if the apparent medication errors in the VAERS database are true errors and if true errors are found, to determine what corrective action was taken locally to prevent or minimize the occurrence of a similar error in the future. Furthermore, if a true error did not occur, we wanted to determine at what point the misinformation was entered into the system.

Methods

The VAERS database was searched for US reports received between July 1, 2001 and June 30, 2002 with either the coding term "medical error," or an entry that had been identified during the initial FDA review of serious reports by medical officers and classified as "error." Our search also looked for the word "error" in the text of a report. In addition, the database was searched for ages of children receiving vaccinations outside of the usual recommendation: measles-mumps-rubella (MMR) 0.0 to 0.9 years, diphtheria-tetanus-acellular pertussis (DTaP) > 8.0 years, diphtheria-tetanus-pertussis (DTP) >8.0 years and diphtheria-tetanus-pertussis-haemophilus (DTPH) > 8.0 years. These search criteria were chosen because the vaccine package inserts specify the proper age of administration. (6,7) The age limits chosen for search purposes were selected to avoid finding vaccinees whose ages were very close to, but outside of, the recommended age range for vaccination.

Adverse event reporters were contacted by telephone to (1) verify information; (2) determine where misinformation had been communicated; (3) learn the circumstances of any medical errors; (4) ask about remedial action if an error had occurred; and (5) to ask about possible sequelae if an error had been made. This study was restricted to domestic reports to facilitate follow-up. To ensure privacy of individual subjects, all data were de-identified.

Results

During the study period, approximately 12,800 domestic VAERS reports were received, of which 5,587 were for the vaccines selected for this study. Follow-up of these reports was concluded in September 2002. A total of 119 reports with apparent inappropriate age at vaccination or with coding terms related to error were found and investigated. Ten reports of inappropriate age and 7 reports of medical error were lost to follow-up. A total of 102 (86%) reporters were contacted successfully; 26 cases (26%) were determined to be actual medication errors, 9 cases (9%) were physician decisions, (a new category), 37 (36%) reports involved data entry errors, and 30 (29%) were reporter errors. These results are shown schematically in Figure 1.

Reporter errors include submitting the wrong date of immunization, wrong vaccine, wrong age at time of vaccination or adverse events inconsistent with the medical record.

Physician decision means that a physician has decided that administering a vaccine outside of the recommendations in the package insert is in the patient's best interest. The types of physician decisions are shown in Table 1. In 6 cases, ages 8 to 41 years, patients presented who required tetanus immunization. The adult vaccine, Td, was unavailable and a childhood vaccine was used instead. DTaP is not recommended at ages greater than 7, as the childhood vaccine contains fivefold more diphtheria toxoid [6]. Two doses of MMR were given one month apart to meet a college entrance requirement. This is in accordance with the Advisory Committee on Immunization Practices recommendation that MMR doses be given at least 1 month apart. (8)

The types and numbers of true errors are shown in Table 2. The most common mistake was wrong inoculum [4]. Six of the wrong inoculum errors involved DTaP/Td/DT confusion. In 5 cases, children from 2 to 10 months of age were given MMR. One person was given Hep B vaccine instead of vitamin [B.sub.12].

The remedial actions taken were usually in-service education and/or retraining. This included instruction to verify prior immunizations in the patient's chart. In another situation, an office changed its procedures so that empty vaccine cartons were discarded immediately, thus preventing vaccine vials from being placed in the "wrong carton." Other actions taken were to segregate vaccines by type in the refrigerator and to purchase vaccines from different manufacturers to decrease look-alike packaging. In one case, a vaccine administrator resigned from her job.

The adverse events reported in conjunction with the reports describing medical errors include cellulitis, seizure, wheezing, weakness, dyspnea, epistaxis, arthralgia, nausea, hypotonia, and food allergy. Except for injection site reactions and possibly fever, there was no clear causal relationship of the possible adverse events to the 26 actual medication errors. The number of cases of DTaP use in adults is too small to make inferences about the adverse event pattern, although 3 of the 6 patients did have injection site reactions of varying degrees.

Discussion

The search strategy used in this study was different from the one used in the previous study of medication errors reported to VAERS. (4) This study included an automated search for apparent inappropriate age of vaccine administration for four vaccines; MMR, DTaP, DTP, and DTPH, that have clearly defined recommended ages for use. Furthermore, the reporters were contacted to determine the correctness of the information received by the FDA and, in the case of true error, what remedial action had been attempted.

The study was limited to VAERS reports within the previous year, so the contact information would likely still be current and the incident would be fresh in the reporter's mind. The most important result of this study is that the true medical errors were only 26% of the apparent medical errors. The errors that creep in during report processing and data entry are predictable, as are the reporter errors. For example, reporters often record current age rather than the age at the time the immunization was given. The new electronic reporting form clearly states which age should be used. (9) Similarity in names such as DTaP and DT can be confusing as well. The percentage of reporter errors may be high in VAERS since many reports are submitted by patients or parents. (10) The physician decision classification is new, to our knowledge. Every medical product should be considered with respect to a particular patient's circumstances. The clinical judgment of one's personal physician may be more appropriate than application of general guidelines for the population as a whole. Shortage of an essential product can lead to some adaptations such as using DTaP in an adult who requires tetanus immunization, as several cases in this study demonstrated. Such adaptation must be done with care since usage in an unfamiliar situation may also lead to errors and adverse effects. (11) The nature of the actual errors found in this study is similar to those reported previously. (4) The new FDA requirement that all medications be bar-coded may help reduce errors associated with wrong inoculums. (12) Incorrect inoculums can be the result of vial confusion; however, bar codes are likely to be used only in larger institutions due to the cost of the required equipment. The Institute for Safe Medication Practices (13) has proposed several steps that could be used to reduce medication confusion. Purchasing products from different manufacturers to reduce look-alike confusion and segregating vials can be accomplished at little or no cost.

[FIGURE 1 OMITTED]

Most errors could have been prevented by more attention to detail. The remedial action taken in the cases found here was usually retraining. Using information technology in drug ordering may help improve patient safety. (14) There were no severe, long-term effects associated with the vaccination errors in our study. Medication errors found in VAERS are certainly a small percentage of the total medication errors made in clinical medicine but they may be representative and worth considering as examples of events that do occur. This study also illustrates the necessity of following up adverse event reports to determine the true facts. (15) Possible vaccine adverse events, including events such as those described in this study, should be reported to the VAERS program (Tel: [phone] or www.vaers.hhs.gov), the pharmaceutical company, or both.

Acknowledgments

We thank all those individuals who submitted reports to VAERS and everyone who participated in this study. Robert Ball, MD made many helpful suggestions. The VAERS working group is: Robert Chen, John Iskander, Elaine Miller, Scott Campbell at the CDC, Ann McMahon, Dale Burwen, Hector Izurieta, Miles Braun, Phillip Perucci at the FDA and Geoffrey Evans at the Health Resources and Services Administration.

References

1. To Err Is Human: Building a Safer Health System. Washington, DC, National Academy Press, 1999.

2. Varricchio F, Iskander J, DeStefano F, et al. Understanding vaccine safety information from the vaccine adverse event reporting system. Pediatr Infect Dis J 2004;23:287-294.

3. Baum C, Kweder SL, Anello C. The spontaneous reporting system in the US. In: Strom BL, ed. Pharmacoepidemiology, 2nd ed. New York, New York, Wiley, 1995, pp 125-155.

4. Varricchio F. Medication errors submitted to the vaccine adverse event reporting system (VAERS). Vaccine 2002;20:3049-3051.

5. Centers for Disease Control and Prevention. Inadvertent intradermal administration of tetanustoxoid-containing vaccines instead of tuberculosis skin tests. MMWR 2004;53:662-664.

6. DTaP, Physician's Desk Reference, 59th ed. Montvale, NJ, Medical Economics Co, 2005, p 756.

7. MMR, Physician's Desk Reference, 59th ed. Montvale, NJ, Medical Economics Co, 2005, p 2075.

8. Centers for Disease Control and Prevention. Measles, mumps, and rubella: vaccine use and strategies for elimination of measles, rubella, and congenital rubella syndrome and control of mumps: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 1998:47:(No. RR-8).

9. Vaccine Adverse Event Reporting System (VAERS) Web site. Available at: http://www.vaers.hhs.gov. Accessed December 5, 2005.

10. Varricchio F. The vaccine adverse event reporting system. J Toxicol Clin Toxicol 1998:36:765-768.

11. Steinmuller DR. A dangerous error in the dilution of 25 percent albumin. N Engl J Med 1998:338:1226-1227.

12. Code of Federal Regulations. 21; [section]201.25; 2004.

13. Institute for Safe Medication Practices. Tetanus toxoid vaccines again used for PPD. ISMP Medication Safety Alert 2004:9:16.

14. Bates DW, Gawande AA. Improving safety with information technology. N Engl J Med 2003;348:2526-2534.

15. Varricchio F. Vaccine adverse event reporting: the importance of follow-up. Expert Rev Vaccines 2005:4:445-448.

Anything's possible if you've got enough nerve.
--J.K. Rowling

Frederick Varricchio, MD, PHD, John Reed, MD, and the VAERS Working Group

From the Division of Epidemiology, Office of Biostatistics and Epidemiology, Center for Biologies Evaluation and Research, Food and Drug Administration, Rockville MD

Reprint requests to Frederick Varricchio, MD PhD, 6130 Roseland Drive, Rockville MD 20852. Email: [e-mail]

This work was done as part of the regular duties of FV as an employee of the US government. JR was an occupational medicine resident at Johns Hopkins Medical School.

The project has been reviewed and approved by the FDA, IRB.

Accepted January 18, 2006.

RELATED ARTICLE: Key Points

* Errors in vaccine administration occur.

* Most errors could be prevented by more attention to detail.

* System changes may be helpful in preventing errors.

* Information in spontaneous reporting should be verified.

Table 1. Physician decision

* DTaP use when Td is unavailable
* Accelerated MMR schedule to meet school requirements
* HIB given at age 4 weeks due to imminent travel
* Age within 1 week of recommended age at regular appointment

MMR, measles-mumps-rubella; DTaP, diphtheria-tetanus-acellular
pertussis; Td, tetanus toxoid; HIB, haemophilus influenza B.

Table 2. True vaccine errors reported to VAERS 2001-2002

Wrong inoculum        13
Overdose               5
Wrong injection site   3
Immunization status    5