The Case

A 2-month-old otherwise healthy infant was admitted to the hospital to rule out sepsis. The infant had been exclusively breastfed since the time of his birth.

In light of the difficulties associated with regular breastfeeding in the hospital, the mother stored her pumped breast milk in the refrigerator on the infant's ward in the hospital. At the time of the first feeding, a nurse's aide retrieved a bottle from the refrigerator, a bottle labeled only with the mother's room number and two initials. Approximately 5 minutes later, after giving the child 15 mL of the breast milk, the mother noticed that the initials on the bottle were not those of her son and called a staff member to the room, who promptly involved the child's main nurse.

The nurse and nurse's aide reviewed the events and suggested to the mother that the administered milk might, in fact, be hers—despite the fact that the initials on the bottle were not those of her child. Upon further review, it was noted that the initials matched those of the nurse's aide who separated the original milk container into small feeding bottles. The mother, skeptical that the milk was hers, asked that the nurses search the refrigerator. There, they found another bottle, labeled with the room number and the child's initials, that clearly was the mother's milk. In other words, they confirmed that the patient had unquestionably received 15 mL of the wrong human milk.

On further investigation, they learned that the mistaken milk sample actually belonged to the patient who previously occupied that room. Although that patient had since been discharged, the breast milk remained in the refrigerator. An infectious diseases consult subsequently counseled the family regarding potential risks from using the wrong breast milk. Additional laboratory testing revealed that no infectious transmission had occurred, and no other objective harm was identified by the time of discharge.

The Commentary

The case reveals a busy pediatric unit; a relatively healthy 2-month-old infant; an astute, insistent mother; a harried nurse's aide; a bottle of breast milk in a refrigerator; and a sequence of system failures. These failures resulted in breast milk being fed to the wrong infant. Of concern is the lack of attention to system flaws that existed prior to this breast-milk error.

Factors Influencing Breast-Milk Errors

A consistent process for handling, storage, and labeling of breast milk was not evident in this case. Standards for infant feed preparation should include the following: an aseptic area or room for transfer and handling of breast milk; accurate labels (preferably computer generated) with clearly defined text provided to the mother at the start of the storage process; and refrigerators that allow storage of a labeled, accessible, and individualized bin for infants' milk.(1) While these implementations address quality assurance and patient safety issues, they do not guarantee elimination of breast-milk errors.

A prevalent health care practice is to "name, blame" and hold accountable the person closest to the error, in this case, the nurse's aide. This "person" approach results in a reactive risk management solution. In this instance, the nurse's aide hypothetically would be re-educated to be more careful and attentive, with the goal of decreasing the possibility of her making another breast-milk error. However, this approach does not prevent another caregiver from making the same error. A more effective approach is proactive management by a multidisciplinary pediatric team to answer the following questions:

• Why were computer-generated breast-milk labels with easy-to-read text not available to the mother and staff?
• Was the breast milk divided into aliquots by the nurse's aide under aseptic conditions?
• Why was the mother bottle feeding when she could have directly breastfed her infant?
• What prompted the aide and the nurse to minimize the mother's concerns regarding the breast-milk label?
• Why was old breast milk allowed to remain in the unit's communal refrigerator?
• Were follow-up care strategies in place when counseling parents after such an error?
• Was the medical testing associated with the error adequate?

While it is a superior mechanism, the proactive solution with a multidisciplinary team is likely to be intensive, lengthy, and costly. This cost can render this approach less appealing for health care facilities.(2)

Practices to Prevent Such Errors May Fail

A commonly used strategy toward preventing breast-milk errors is visual verification. Two RNs or an RN and another person visually and/or verbally verify the accuracy of the breast-milk label, which is generally followed by double signatures. However, in an effort to improve efficiency, staff members tend to work around tasks perceived to be time consuming, repetitive, and seemingly uncomplicated, such as infant feedings.(3) Experienced, knowledgeable, and dedicated staff nurses have candidly confided that the diligent double-check can evolve into a cursory glance, with a presumption that one's colleagues' work must be accurate. Second, visual verification can be affected by an interesting phenomenon known as "inattentional blindness." Despite viewing and checking a label, one sees what one expects to see, not what is actually written. This phenomenon is the "natural result of two fundamental aspects of the human condition: limited mental attention and high adaptability."(3) This inattentional blindness often results in unreliability of visual verification for infant feeds.

Frequency of Breast-Milk Errors

One pediatric hospital calculated the risk for breast-milk errors to be 1.04 errors per month, based on an average of 10,000 monthly feeds.(4) A study of a 42-bed neonatal intensive care unit (NICU) documented 80 breast-milk errors over the course of 10 years.(5) Of those, approximately 99% took place in infants whose surnames had little or no similarities, while only 1% were associated with infants with the same surname. Despite various label changes and visual verification approaches, breast-milk errors continued to occur.(5) Another study noted that critical incident monitoring resulted in reporting of breast-milk errors, but it did not prevent them from occurring.(6)

Risk of Exposure to and Disease Transmission from the Wrong Breast Milk

In this case, the 8-week-old relatively healthy infant who ingested 15 mL of the wrong breast milk would be at negligible risk to develop disease. Human immunodeficiency virus (HIV) can pass from mother to child via breast milk. The probability for transmission from an HIV-positive mother to an infant through breastfeeding was analyzed to be 0.00064 per liter of breast milk ingested and 0.00028 per day of breastfeeding.(7) Developed countries practice routine prenatal screening for HIV, unless the mother opts out. HIV-positive mothers are not allowed to breastfeed or bottle feed their expressed breast milk, thereby reducing the risk of transmission.(8) However minute the risk, it does little to ease the angst of parents when discussing the possibility that their infant was inadvertently exposed to HIV. Human T-cell leukemia virus type I (HTLV-I) may be a cause of adult T-cell leukemia and may be transmitted in mothers' breast milk.(9) In developed countries, it is contraindicated for mothers with HIV or HTLV to breastfeed or use their breast milk.(10,11) A mother positive for hepatitis B (HBV) surface antigen has little risk of transferring the virus to her own infant. Nonetheless, the recommended practice is to provide immunoglobulin and HBV vaccine after birth to the infant, and breastfeeding can begin after hepatitis B vaccination. Hepatitis C, while found in breast milk, is unlikely to be transmitted; therefore, testing is not required. Cytomegalovirus, also in found in breast milk, has little impact on healthy full-term or older infants, and routine testing is not recommended.(8,12)

Follow-Up Care for Breast-Milk Errors

The case investigation alluded to laboratory testing, family counseling by expert staff, and no evidence of harm by time of discharge. This assumes contact had been made with the donor mother and that she was tested for HIV, HTLV, and HBV and was found not to be a carrier. However, follow-up care of the 8-week-old infant is more pressing if the donor mother had been HBV surface antigen–positive or if she had refused testing. In either case, the infant should receive immunization for HBV.(8) Seroconversion for HIV antibodies generally occurs within 6 weeks of exposure, and the infant should have blood testing to determine serologic status at that time.(13)

Breast-milk errors are infrequent, making care and counseling a challenge for staff. An important component of proactive management is being prepared for the worst. To promote a consistent approach by the physician, nurse, infection control practitioner, and infectious disease practitioner, the use of pre-packaged breast-milk error kits can prevent important steps from being missed. The package should contain information pamphlets for recipient and donor mother/parents explaining the viruses of concern, risk of disease transmission, and consent forms for blood testing. Medical staff should have a detailed list of blood tests to be performed, blood volume samples, who should be notified of the error, and steps to be taken in the event that a donor mother refuses testing.(8,14)

Preventing Breast-Milk Errors in a Hospital Setting

Actual breastfeeding is certainly the best deterrent to breast-milk errors. However, in instances when breastfeeding is impossible and stored milk is used, wireless bar code scanning eliminates the risk of inattentional blindness and provides efficient verification methods and ongoing data collection. Had bar coding been available in this hospital, unique identifiers for this infant would have been scanned and revealed a mismatch with the bar code label on the bottle of breast milk. The nurse's aide would have been alerted to the breast milk mismatch, preventing the error. Technology has the ability to compensate for the environment (hectic assignments and human shortcomings) by preventing breast-milk errors independent of the caregiver's experience, training, or label interpretation.(5)

In a recent article, the use of a bar code scanning system decreased breast-milk errors from a yearly average of 8 to 0. There were 237 audio alert alarms for mismatched milk and patient (same or similar surnames, and surnames with no similarities) in the first year of use. A bar code embedded with 4 unique identifiers (infant surname, medical chart number, date of birth, and gender) was generated at the time of the infant's admission. Subsequently, all other bar code labels for milk collection, feed orders, feed preparation, and feeding were created. This sequential system requires correct performance of step A before being allowed to move to step B, reducing workarounds and providing audio safeguards to alert staff to a mismatch of breast milk and patient.(5)

Interestingly, the NICU using the point-of-care bar code scanners recently had two breast-milk error events, the first since February 2008. In both cases, the breast-milk container was scanned by the nurse during, and not before, the feed (as the protocol calls for). It was only after the feed had begun that the RNs were alerted to the breast-milk error by the audio alarm. This demonstrates that the bar coding system works, but that human workers can decrease the benefit of reliable technological systems if the processes and culture are not strong.(15)

Take-Home Points

• Encourage direct breastfeeding whenever possible.
• Engage mothers to validate their own milk and do not dismiss their concerns; they are likely to be the most conscientious caregivers when it comes to their own infants.
• Develop bedside point-of-care bar code scanning, and support or champion other system improvements that decrease the reliance on human perfection to get important tasks right all the time.
• Shift the paradigm from reactive risk management to proactive management. Be prepared for a breast-milk error, as preparedness will make the error more manageable.
• The risk of high parental anxiety is much greater than the risk of disease transmission from a breast-milk error; however, the risk is not zero.
• Be efficient and organized in providing follow-up care after a breast-milk error. Promptly report results to parents.

Dorothy Dougherty, RN

Neonatal Intensive Care Unit

Sunnybrook Health Sciences Centre

Toronto, Canada

References

1. Robbins ST, Beker LT, Pediatric Nutrition Practice Group. Infant Feedings: Guidelines for Preparation of Formula and Breastmilk in Health Care Facilities. Chicago, IL: American Dietetic Association; 2004. ISBN: 9780880913096.

2. Reason J. Human error: models and management. BMJ. 2000;320:768-770. [go to PubMed]

3. Green M. Nursing error and human nature. J Nurs Law. 2004;9:37-44.

4. Drenckpohl D, Bowers L, Cooper H. Use of the six sigma methodology to reduce incidence of breast milk administration errors in the NICU. Neonatal Netw. 2007;26:161-166. [go to PubMed]

5. Dougherty D, Nash A. Bar coding from breast to baby: a comprehensive breast milk management system for the NICU. Neonatal Netw. 2009;28:321-328. [go to PubMed]

6. Zeilhofer UB, Frey B, Zandee J, Bernet V. The role of critical incident monitoring in detection and prevention of human breast milk confusions. Eur J Pediatr. 2009;168:1277-1279. [go to PubMed]

7. John-Stewart G, Mbori-Ngacha D, Ekpini R, et al; Ghent IAS Working Group on HIV in Women Children. Breast-feeding and transmission of HIV-1. J Acquir Immune Defic Syndr. 2004;35:196-202. [go to PubMed]

8. Warner BB, Sapsford A. Misappropriated human milk: fantasy, fear, and fact regarding infectious risk. Newborn Infant Nurs Rev. 2004;4:56-61. [Available at]

9. Takeuchi H, Takahashi M, Norose Y, Takeshita T, Fukunaga Y, Takahashi H. Transformation of breast milk macrophages by HTLV-I: implications for HTLV-I transmission via breastfeeding. Biomed Res. 2010;31:53-61. [go to PubMed]

10. MacDonald NE. Maternal infectious diseases, antimicrobial therapy or immunizations: very few contraindications to breastfeeding. Paediatr Child Health. 2006;11:489-491. [Available at]

11. Gartner LM, Morton J, Lawrence RA, et al; American Academy of Pediatrics Section on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics. 2005;115:496-506. [go to PubMed]

12. Monif GRG, Baker DA, ed. Infectious Diseases in Obstetrics and Gynecology, 6th edition. NY, NY: Informa Healthcare; 2008. ISBN: 9781842142097.

13. American Academy of Pediatrics, American College of Obstetricians and Gynecologists. Human immunodeficiency virus screening. Joint statement of the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists. Pediatrics. 1999;104:128. [go to PubMed]

14. Dougherty D, Giles V. From breast to baby: quality assurance for breast milk management. Neonatal Netw. 2000;19:21-25. [go to PubMed]

15. McDonald CJ. Computerization can create safety hazards: a bar-coding near miss. Ann Intern Med. 2006;144:510-516. [go to PubMed]