The childhood vaccination program recommended by the Centers for Disease Control and Prevention (CDC) has resulted in a dramatic reduction in the number of cases of measles (rubeola), mumps, and rubella infections and congenital rubella syndrome.1 As part of a nationwide effort to eliminate congenital rubella syndrome, reproductive-aged women are screened for immunity to the virus in prenatal clinics. In addition to childhood vaccines, persons born after 1957 who are aged 18 years or older who cannot document that they have received at least 1 dose of a measles-, rubella- and mumps-containing vaccine (MMR) or that they have immunity to these 3 viruses are reimmunized with a “booster” injection. Despite aggressive vaccination strategies, however, susceptibility to rubella, rubeola, and mumps viruses remain. Susceptibility arises from several factors, including failure of proper childhood vaccination or faded immunity.1–4 Estimates of susceptibility to rubella infection in reproductive-aged women in the United States (U.S.) range from 10–18%1,4–6 but may be more than 40% in other countries.7 Additionally, 1 study demonstrated a 20.7% seronegativity rate for measles and 15.6% for mumps in U.S. military recruits.8
The American College of Obstetricians and Gynecologists (ACOG) currently recommends screening pregnant women only for immunity to rubella infections.9 Those who are susceptible to rubella are to receive an MMR vaccine after delivery. However, if susceptibilities to mumps and rubeola are significant in women with immunity to rubella, obstetricians will miss an opportunity to vaccinate women still susceptible to those infections. The objective of this study was to estimate rubella, rubeola, and mumps susceptibilities to assess the percentage of women who were not immune to rubeola or mumps, depending on rubella immunity status. A secondary objective was to assess costs of vaccination and testing programs aimed at eliminating these viral susceptibilities to determine an optimal strategy.
MATERIALS AND METHODS
We performed a cross-sectional, observational study of women receiving care in the prenatal clinics at Naval Hospital Camp Lejeune. The Obstetrics Department at Naval Hospital Camp Lejeune provides obstetric care for active-duty members and military dependents of Marine Corps Base Camp Lejeune and the surrounding area. The Naval Hospital is a 117-bed facility that averages 1,500 deliveries per year. The base population encompasses more than 40,000 active-duty Navy and Marines and approximately 95,000 family members and retiree beneficiaries. Before the study period, all women presenting for care gave blood samples to determine rubella seropositivity along with their other prenatal laboratory tests.
Upon presenting for initial obstetric care, our patients attend an initial screening nurse visit where the prenatal laboratory studies are ordered and reviewed. At that visit, the MMR titers were obtained with the remainder of the prenatal labs. In May 2004, our facility began to obtain viral immunoglobulin G (IgG) titers for immunity to rubella, rubeola, and mumps on all women presenting for initial prenatal care. The enzyme-linked immunosorbent assay (Wampole Laboratories, Princeton, NJ) was used for this test. The test was performed at Armstrong Laboratories at Brooks Air Force Base, TX. A level of 1.1 or greater represented immunity. The patient’s age, gravidity, parity, and immunity status to these viruses were recorded for this study. Gravidity was separated into primigravid and multigravid women. Parity was divided into nulliparous and parous groups. This was done under the assumption that women who had been pregnant in the past may have been tested for rubella susceptibility and possibly immunized if susceptible whether their pregnancy ended in miscarriage, termination, or birth. Additionally, the women were asked if they received an MMR “booster” vaccine in their late teenage or early adult years. Women who miscarried before they were asked this question, or if an answer had not been given, were placed into the “unsure/unknown” category. This study represents the collected data from all women presenting for initial prenatal care from May 2004 through October 2004.
The data were placed in a Microsoft Access (Microsoft Corp., Redmond, WA) database and analyzed using SPSS 12 (SPSS Inc., Chicago, IL). Results were compared using χ2 and t tests. A logistic regression was then performed to control for confounding and determine any predictor variables for immunity to all 3 viruses. The initial model included the variables age, multigravid compared with primigravid, parous compared with nulliparous, and knowledge of receiving a booster vaccine all entered in a block. Backward stepwise regression selected the final model.
At Naval Hospital Camp Lejeune, the vaccine cost for a rubella vaccine is $10.08 and $25.51 for an MMR (personal communication, Naval Hospital Camp Lejeune pharmacy). The average retail cost for each of these vaccines is $14.87 (personal communication, Merck pharmaceuticals) and $38.05 (www.cdc.gov/nip/vfc/cdc_vac_price_list.htm) respectively. The laboratory cost of performing a rubella virus IgG for seropositivity is $4.00 and for susceptibility to all 3 viruses is $15.00 (personal communication, Naval Hospital Camp Lejeune laboratory). These laboratory costs and the retail vaccine costs were used to develop a cost comparison of testing and vaccine strategies, based on our seronegativity results, for 1,000 patients.
The hospital’s governing institutional review board approved this project. Because the project used data collected as part of routine prenatal care, informed consent was not required.
In the 6 months of the study, 973 pregnant women presented for initial obstetric care. All of them had an MMR titer drawn. The mean age of the women was 24.2 years (range 17–43 years). There were 358 primigravidas (36.8%; gravidity range 1–13), and 485 (49.8%) were nulliparous (parity range 0–7). Overall, 91 (9.4%) women were susceptible to rubella, 161 (16.5%) to rubeola, and 159 (16.3%) to mumps. Three hundred seventeen (32.6%) were susceptible to at least 1 virus, whereas only 17 (1.7%) were not immune to all 3. More than one half of the women who were not immune to rubella were also susceptible to either rubeola or mumps (n = 47, 51.6%). Women who were immune to rubella also had a large susceptibility to either rubeola or mumps (n = 226, 25.6%).
Table 1 represents the data comparing those women with viral susceptibilities to those found immune. The only statistically significant difference found was that multigravid women were more often immune to mumps.
The majority of women (n = 733, 75.3%) did not remember whether they had received a booster vaccination (hereafter “unknown”), whereas 229 (23.5%) noted that they did receive the booster vaccination. Eleven women (1.1%) specifically remembered not receiving a booster vaccine. Of those women not receiving a vaccine, 54.5% were immune to rubella, 72.7% were rubeola immune, and 72.7% were immune to the mumps. Only 27.3% were immune to all 3 viruses, compared with 74.2% of those who responded “yes” to receiving the booster, and 65.9% of those in the unsure group (P = .001). Because the group of women who responded “no” represented only 1.1% of the sampled group, their responses were combined with the “unknown” group for subsequent analyses. This was done to facilitate a worst-case scenario of all of the women in the unsure group not having received a booster. The comparison between those who did receive an MMR booster and those who were in the other group are displayed in Table 2.
A logistic regression model was used to assess for predictors of immunity to all 3 viruses and also to predict booster status. Table 3 represents these models. Knowing that a booster was given predicts immunity to all 3 viruses. A higher age predicts not knowing whether a pregnant woman received a booster.
Using the above percentages for immune status, we modeled screening and vaccine strategies for 1,000 patients. In our study this is roughly equivalent to costs for 6 months of patients. The strategies are Strategy A—test for rubella susceptibility and if nonimmune, give a rubella-only vaccine; Strategy B—test for rubella susceptibility and if not immune give an MMR vaccine (current ACOG recommendation); Strategy C—test for susceptibilities to entire MMR panel and if not immune to 1, give an MMR vaccine; Strategy D—test for susceptibilities to entire MMR panel and if susceptible to rubella alone give rubella-only vaccine but give MMR for susceptibility involving rubeola or mumps. The cost distribution for the 4 plans is represented in Table 4. Strategy A would not immunize 280 women who were susceptible to a virus (48 women who were also not immune to rubeola or mumps plus 232 who were immune to rubella but not immune to either rubeola or mumps). Strategy B would not immunize 232 women who were immune to rubella but susceptible to either rubeola or mumps. Strategies C and D would immunize all women who were susceptible to 1 of the 3 viruses.
Our study demonstrates that in 2004, one third of our pregnant women lack immunity to either rubella, rubeola, or mumps. Women who are immune to rubella often lack immunity to the other components of the MMR vaccine. This finding indicates that the current pregnancy population screening and vaccination strategy may be inadequate. Even those pregnant women who knew they had received a booster vaccine failed to demonstrate immunity to all components of the MMR in 25.8% of the cases. Although receiving the booster vaccine does significantly improve a woman’s chances of having immunity to all 3 viruses, the amount of susceptibility in our patient population is surprising and similar to rates published from older data.1,4–6 Although indigenous transmission in the United States of rubeola may have been interrupted in 1993,10 immunity to that virus is important, because the virus may be imported from other countries. Measles infections can be complicated by encephalitis and death. Measles infection in pregnancy increases rates of premature labor, miscarriage, and low birth weight.1 Additionally, although congenital rubella syndrome cases have decreased, 65% of the reported cases of rubella occur in persons aged 20 years or older.1 The incidence of mumps is very low in the United States, but the most serious complications of infection, including orchitis, meningitis, and meningoencephalitis, occur among adults infected with the virus.1 Although progress at eliminating these viral diseases in this country has been made, a large percentage of the child-bearing population is still susceptible.
Data from previous studies demonstrate that postpartum vaccination programs can be successful at reducing rubella susceptibility in pregnant seronegative women.1,4,11 However, a recent survey of U.S. hospitals revealed that only 21% of the hospitals offering labor and delivery services had implemented a rubella vaccination program for postpartum women.12 The CDC1 and ACOG13 recommend postpartum vaccination for susceptible women. The CDC recommends that women of child bearing age who are susceptible to rubella receive a rubella-containing vaccine (ie, rubella, measles plus rubella, or MMR).1 ACOG also specifies the use of an MMR vaccine for postpartum women; however, this recommendation was based on a rubella-only vaccine shortage.13 This shortage may no longer be present. The MMR vaccine has very few adverse effects and has been found to be generally safe.1,14
Few studies have looked at the relation between rubella susceptibility and that to rubeola and mumps in pregnant women. In Canada, Libman et al2 demonstrated that 8.2% of women seronegative to rubella were also seronegative to measles, compared with only 0.8% susceptible if the woman was rubella seropositive. In addition, they also found a 7.8% and 7.4% susceptibility to mumps if the woman was rubella seronegative or seropositive respectively. The Canadian rubeola data are in contrast to Coupland et al,3 who found similar distributions of viral antibody to measles regardless of rubella immune status in the United Kingdom. Military surveys of susceptibility in the United States show a modest association between rubella and rubeola susceptibility, but did not study pregnant women specifically and were published before the outbreaks of measles and rubella in the early 1990s.1,8,15
The difference we found in age with respect to receiving a booster is likely due to the fact that, in general, our active-duty population tends to be younger and have a higher rate of receiving a booster than dependents, who generally were older. We did not link the records with active-duty status, however, so this is speculation. Our study also pointed out that many women did not remember whether they received a booster vaccine. This finding mirrors that of Preblud et al15 that about 75% of reproductive-aged cadets did not know their vaccination status. As health care providers, we must better educate our patients about interventions they receive so they remember that they had them when asked later in life.
A strength of our study is the large number of pregnant women studied. Other studies we found were limited to fewer patients.2,3 Our study examines viral susceptibilities of U.S. active-duty and dependent pregnant women, many of whom were in their adolescent years during the main focus of immunization programs in this country. Other studies of U.S. service members had a minority of female subjects (22–26%) and reported susceptibility rates before 1990.6,8 We found 1 other study linking the viral immunities to self-recall of booster status. That study, however, reported on a population in 1977 that was 96.5% male.15 Also, our study was cross-sectional in nature and all women had the same objective measures. Self-recall of booster vaccine remains the only variable subjective in nature. This limits ascertainment bias in the study.
One weakness of the study is that ethnicity was not measured. It is known that the rubella outbreaks in the 1990s occurred primarily among persons of Hispanic origin.1 The patient population of our obstetric clinic is ethnically diverse, but this variable was not ascertained as part of this study. Our patients come from all geographic regions of the United States. Thus, we think that our results are generalizable. Recall bias could have been introduced by asking patients their vaccine booster status. We did not ask patients whether they had actually had 1 of these viral illnesses in the past.
In assessing which screening and vaccination strategy to implement, health care providers must focus on the goal of these programs. The program currently recommended by ACOG is strategy B. This seems a cost-effective strategy, but if our findings are generalizable to the U.S. population, a significant proportion of patients will miss an opportunity for vaccination to illnesses to which they are susceptible. Although it is more costly to implement a strategy such as C, it allows all women with susceptibilities to MMR to obtain the needed vaccine booster. We must examine our role in obstetrics as women’s health care physicians. If our goal is to ensure maximal vaccination of all susceptible women, then Strategy C should be implemented. If, however, we rely on low prevalence of rubeola and mumps infections, then we can continue testing only for rubella immunity as in Strategy B. Pregnancy is a time when health care providers can intervene for their patients to improve their overall health. Screening for viral illness susceptibility during antenatal care and subsequent vaccination of those found seronegative is a way providers can help promote population health.
In conclusion, many pregnant women are susceptible to rubella, rubeola, and mumps viruses when they present for prenatal care. A quarter of the women immune to rubella were susceptible to either rubeola or mumps. The majority of women do not remember whether they received a vaccine booster. Having received a booster predicts immunity to all 3 viral components of the MMR vaccine. Women’s health care providers must continue to work to eradicate these viral susceptibilities in reproductive-aged women.
1. Watson JC, Hadler SC, Dykewicz CA, Reef S, Phillips L. 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 Recomm Rep 1998;47(RR-8):1–57.
2. Libman MD, Behr MA, Martel N, Ward BJ. Rubella susceptibility predicts measles susceptibility: implications for postpartum immunization. Clinic Infect Dis 2000;31:1501–3.
3. Coupland B, Revill S, Desselberger U. Relation between susceptibility to rubellavirus infection in pregnant women and decreased immune responsiveness. Vaccine 1990;8:190–1.
4. Griffiths PD, Baboonian C. Is post partum rubella vaccination worthwhile? J Clin Pathol 1982;35:1340–4.
5. McElhaney RD Jr., Ringer M, DeHart DJ, Vasilenko P. Rubella immunity in a cohort of pregnant women. Infect Control Hosp Epidemiol 1999;20:64–6.
6. Crawford GE, Gremillion DH. Epidemic measles and rubella in Air Force recruits: impact of immunization. J Infect Dis 1981;144:403–10.
7. Yadav S, Gupta S, Kumari S. Seroprevalence of rubella in women of reproductive age. Indian J Pathol Microbiol 1995;38:139–42.
8. Kelley PW, Petruccelli BP, Stehr-Green P, Erickson RL, Mason CJ. The susceptibility of young adult Americans to vaccine-preventable infections. A national serosurvey of US Army recruits. JAMA 1991;266:2724–9.
9. American Academy of Pediatrics, American College of Obstetricians and Gynecologists. Guidelines for perinatal care. 5th ed. Elk Grove Village (IL): AAP; Washington, DC: ACOG; 2002.
10. Watson JC, Redd SC, Rhodes PH, Hadler SC. The interruption of transmission of indigenous measles in the United States during 1993. Pediatr Infect Dis J 1998;17:363–6.
11. Bottiger M, Forsgren M. Twenty years’ experience of rubella vaccination in Sweden: 10 years of selective vaccination (of 12-year-old girls and of women postpartum) and 13 years of a general two-dose vaccination. Vaccine 1997;15:1538–44.
12. Bath SK, Singleton JA, Strikas RA, Stevenson JM, McDonald LL, Williams WW. Performance of US hospitals on recommended screening and immunization practices for pregnant and postpartum women. Am J Infect Control 2000;28:327–32.
13. Rubella vaccination. ACOG Committee Opinion No. 281. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;100:1417
14. Smeeth L, Cook C, Fombonne E, Heavey L, Rodrigues LC, Smith PG, Hall, AJ. MMR vaccination and pervasive developmental disorders: a case-control study. Lancet 2004;364:963–9.
15. Preblud SR, Gross F, Halsey NA, Hinman AR, Herrmann KL, Koplan JP. Assessment of susceptibility to measles and rubella. JAMA 1982;247:1134–7.