Sick and Pregnant
A 35-year-old woman with chronic asthma presented to the emergency department (ED) with difficulty breathing. The patient informed the staff that she was 17 weeks pregnant and had an obstetrician on staff at another hospital. A urine pregnancy test was ordered and was positive. The test result was documented in the electronic medical record. The patient was treated with inhaled bronchodilators, but her respiratory distress persisted. The ED physician contacted the on-call internist to admit the patient for continued therapy. The internist agreed to admit the patient, but he was not told that the patient was pregnant. The admitting nurse received a report from the ED nurse, but again, the patient's pregnancy status was not mentioned.
On admission, the patient was ordered to receive intravenous corticosteroids, nebulized bronchodilators, and intravenous levofloxacin (a pregnancy category C antibiotic [see Table]). In the morning, the internist saw the patient. She informed him that she was pregnant. The physician reviewed the patient's medication administration record and determined that she received one dose of levofloxacin. He discontinued levofloxacin and ordered an alternate antibiotic that was pregnancy category B. A maternal-fetal specialist was consulted and reported that one dose of levofloxacin should have no adverse effects on the fetus.
This case brings up several issues related to medication safety in women who are, or could be, pregnant—including problems in interprofessional communication, patient education, and medication-use systems. Grissinger and Kelly outlined some of the causes of medication errors in women and recommended strategies to improve medication safety in this unique group.(1) The safe and effective use of medications begins with an accurate medical/medication history. In caring for women, such a history should include height, weight, current and past medical history (including current pregnancy, past pregnancy, and lactation), as well as a comprehensive medication history that includes prescription, over-the-counter, and complementary/alternative medicine products.(1) Drug use by pregnant women is more common than one might expect. In two studies, nearly 60% of pregnant women used a prescription medicine other than a prenatal vitamin. In many cases, these prescription drugs were not prescribed by an obstetrician—the provider group most likely to be knowledgeable about drug safety during pregnancy.(2,3) Also, key elements of an accurate medication history are often missing in the health records of pregnant women. In one study, only 26% of medications taken by women during pregnancy were actually documented in the patient's chart.(4) Medication errors related to sound-alike, look-alike drugs (SALAD) can occur. Examples of SALAD associated with medication errors in women include the mix-up of Sarafem (fluoxetine) for Serophene (clomiphene) and the inadvertent substitution of FemHRT (a hormone replacement product used for vasomotor symptoms during menopause) for Femara (letrozole, used to prevent cancer progression following antiestrogen therapy).(1)
To help practitioners weigh the risks and benefits of drug exposure during pregnancy, the United States Food and Drug Administration (FDA) in 1979 devised a categorization scheme to communicate the safety of medications in pregnancy in terms of effects on the fetus (Table). However, this system has been widely criticized for its lack of clarity and lack of a strong evidence basis.(5-7) For example, categories do not clearly communicate relative risk (e.g., risk does not necessarily increase as one goes from categories C to D to X) nor how the risk of drug exposure might vary depending on the trimester of pregnancy during which the exposure occurs. Some drugs, such as oral contraceptives, are classified as category X, when data show that they are not as detrimental to the fetus as category X suggests.(8)
Responding to these critiques, in May 2008 the FDA proposed a new rule to remove the pregnancy categories and instead require that product labels include three main components: a fetal risk summary, clinical considerations, and a data section in the labeling.(9) Information in these three areas would include the developmental risk to the fetus, specific patient counseling and prescribing information, and data from studies in humans and animals (if available). Safety information for exposure during various stages of pregnancy and during lactation would be included, as would information related to pregnancy exposure registries. The FDA is presently reviewing and taking into consideration comments submitted between May 28, 2008 and August 27, 2008 for modification of this proposed rule.
It is hard to know how knowledgeable prescribers are about pregnancy status when medications are initiated or refilled. However, as this case illustrates, there does seem to be cause for concern. In 2004, investigators reviewed pregnancies from eight HMOs between 1996–2000 and found that, even though pregnancy was documented in the medical chart, 3.8% of women were prescribed category D medications and 1.1% were prescribed category X medications.(10)
While obtaining medical histories, most providers routinely ask if a woman is, or could be, pregnant. Most EDs have policies in place that require providers to ask about pregnancy status or to order pregnancy tests before proceeding with certain medical treatments. In this case, the patient informed the staff that she was pregnant, and a pregnancy test was performed; however, the error occurred in the communication of this information between providers. It seems that systems were not in place along the continuum of care to ensure that the patient's pregnancy status was known by all of the relevant care providers.
Improved provider-to-provider communication and up-to-date software could have prevented the information gap that led to this error. The Joint Commission's 2008 National Patient Safety Goals (NPSGs) include improving communication between caregivers. Communication issues among the health care team are the subject of improvement projects and interprofessional training.(11) Let's review the points of care where the error could have been prevented. First, if the ED nurse and physician had simply stated the pregnancy status as part of the transfer of care (e.g., as part of a structured sign-out procedure), it might have been enough. Second, the internist could have reviewed the information more carefully in the medical record. Third, the nurse administering the medication could have verified the patient's pregnancy by asking the patient or reviewing the chart. The pharmacist also could have inquired if the patient was pregnant before dispensing the medication. It is unknown if the pharmacy computer system at this hospital was linked to the electronic medical record or had flags to alert the pharmacist of pregnancy status. A randomized study examined the impact of embedding a pregnancy alert tool into an electronic pharmacy system.(12) The study showed a significant reduction in the dispensing of category D or X medications to pregnant women, demonstrating that a pharmacy alert system may prevent these errors. Despite the positive results, the study's authors cautioned that the alert system alone was not enough to prevent errors and that changes in behavior by both physicians and pharmacists, in conjunction with systems changes, were required to achieve optimal outcomes. Lastly, another opportunity for improved communication involves the active participation of the patient herself. Among the 2008 NPSGs, the Joint Commission encourages changes to allow patients to play an active role in their care.(13) This patient could have been educated (e.g., by her obstetrician, her pharmacist, or by an admitting health care provider) to mention her pregnancy to providers at each point of contact. In this way, the error may also have been prevented.
A computerized provider order entry (CPOE) system, with adequate decision support, may have also prevented this situation. With such a system, a flag may have appeared to alert the physician that levofloxacin is not the best choice of antibiotic for a pregnant woman. This, however, would not guarantee that the error would have been prevented. Data show that health care team involvement, in addition to CPOE, is important to preventing medication errors because health care providers often override electronic alerts.(14,15)
In the end, it was fortunate that the error only resulted in a single dose of levofloxacin, which is highly unlikely to be harmful to the patient or the fetus. But other medications may not have been as forgiving. It is important to learn from the errors and points of communication loss in this case to help avoid similar mistakes in the future.
- All health care providers (physicians, pharmacists, nurses, dentists, etc.) should develop the practice of verifying if a female patient is pregnant.
- Clear sign-out procedures in the transfer of care and patient education are critical for adequate communication across the continuum of care.
- CPOE and pharmacy alert systems can alert providers to the pregnancy status of patients and suggest that alternatives be considered when certain high-risk drugs are prescribed for pregnant patients. However, these efforts do not replace the need for interprofessional education and communication.
- The current FDA categories for medication use in pregnancy may not fit all of health care providers' informational needs. A new FDA proposal, currently under consideration, would revise drug product labeling related to pregnancy and lactation with the goal of providing information that is more useful for clinical decision-making and for patient counseling.
- Pregnant women should be encouraged to take an active role in educating all health providers, at each point of contact, about their pregnancy status.
Shareen El-Ibiary, PharmD, BCPS
Associate Professor of Clinical Pharmacy
University of California, San Francisco
1. Grissinger MC, Kelly K. Reducing the risk of medication errors in women. J Womens Health (Larchmt). 2005;14:61-67. [go to PubMed]
2. Glover DD, Amonkar M, Rybeck BF, Tracy TS. Prescription, over-the-counter, and herbal medicine use in a rural, obstetric population. Am J Obstet Gynecol. 2003;188:1039-1045. [go to PubMed]
3. Glover DD, Larsen B. Longitudinal investigation of candida vaginitis in pregnancy: role of superimposed antibiotic use. Obstet Gynecol. 1998;91:115-118. [go to PubMed]
4. Bodendorfer TW, Briggs GG, Gunning JE. Obtaining drug exposure histories during pregnancy. Am J Obstet Gynecol. 1979;135:490-494. [go to PubMed]
5. Boothby LA, Doering PL. FDA labeling system for drugs in pregnancy. Ann Pharmacother. 2001;35:1485-1489. [go to PubMed]
6. Addis A, Sharabi S, Bonati M. Risk classification systems for drug use during pregnancy: are they a reliable source of information? Drug Saf. 2000;23:245-253. [go to PubMed]
7. FDA classification of drugs for teratogenic risk. Teratology Society Public Affairs Committee. Teratology. 1994;49:446-447. [go to PubMed]
8. Scialli AR. The perils of the pregnancy label. Reprotox. Bethesda, MD: Reproductive Toxicology Center; 2004. [Available at]
9. Summary of Proposed Rule on Pregnancy and Lactation Labeling. Rockville, MD: Center for Drug Evaluation and Research, US Food and Drug Administration. [Available at]
10. Andrade SE, Gurwitz JH, Davis RL, et al. Prescription drug use in pregnancy. Am J Obstet Gynecol. 2004;191:398-407. [go to PubMed]
11. Varpio L, Hall P, Lingard L, Schryer CF. Interprofessional communication and medical error: a reframing of research questions and approaches. Acad Med. 2008;83:S76-S81. [go to PubMed]
12. Raebel MA, Carroll NM, Kelleher JA, et al. Randomized trial to improve prescribing safety during pregnancy. J Am Med Inform Assoc. 2007;14:440-450. [go to PubMed]
13. 2009 National Patient Safety Goals. Joint Commission on Accreditation of Healthcare Organizations. [Available at]
14. Koppel R. Computerized physician order entry systems: the right prescription? LDI Issue Brief. 2005;10:1-4. [go to PubMed]
15. Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry systems in facilitating medication errors. JAMA. 2005;293:1197-1203. [go to PubMed]
Table. The FDA's present system to categorize the safety of medications in pregnancy, in terms of the risk to the fetus.
Controlled studies show no risk. Adequate, well-controlled studies in pregnant women have failed to demonstrate risk to the fetus.
No evidence of risk in humans: either animal studies show risks, but human studies do not; or no human studies have been done and no risk has been shown in animal studies.
Risk cannot be ruled out. Human studies are lacking, and animal studies are either positive for fetal risk or lacking as well. However, potential benefits may justify potential risk.
Positive evidence of risk. Investigational or postmarketing data show risk to the fetus. Nevertheless, potential benefits may outweigh the potential risk.
Contraindicated in pregnancy. Studies in animals or humans, or investigational or postmarketing reports, have shown fetal risk, which clearly outweighs any possible benefit to the patient.
ethinyl estradiol, methotrexate, isotretinoin