Allsworth, Jenifer E. PhD1; Peipert, Jeffrey F. MD, MPH1
Sexually transmitted diseases (STDs), including infection with the human immunodeficiency virus (HIV), are important and costly public health problems in the United States. Women with bacterial vaginosis may be at increased risks for STDs, HIV, and a number of other adverse reproductive outcomes, including pelvic inflammatory disease (PID), postoperative infections, spontaneous abortion, preterm birth, and postpartum endometritis. Specifically, bacterial vaginosis has been found to be associated with the prevalence and incidence of multiple STDs, including chlamydia, gonorrhea, herpes, HIV, and trichomoniasis, and is also implicated in the development of PID.1–8
Bacterial vaginosis is a highly prevalent condition and the most common cause of vaginal irritation. It is a condition characterized by vaginal flora imbalance, in which normally plentiful peroxidase-producing bacteria (Lactobacillus) are scarce and other anaerobic bacteria abundant. The total concentration of bacteria may be 100 to 1,000 times their normal levels in women with bacterial vaginosis.9 Many, if not most, women with bacterial vaginosis are asymptomatic. The two classic symptoms of bacterial vaginosis, discharge and odor, are reported by only a minority of affected individuals. A recent study found that, in women with bacterial vaginosis, only 25% reported odor and 42% reported discharge in the preceding 6 months.10
Although bacterial vaginosis is a common condition, national surveillance has been lacking. Data from the 2001–2002 National Health and Nutrition Examination Survey found bacterial vaginosis to be common among the general population (Koumans EH, Sternberg MR, McQuillan G, Bruce C, Kendrick JS, Sutton MY, et al. Prevalence of bacterial vaginosis in the United States, 2001–2002. Presented at the 2006 National STD Prevention Conference. Jacksonville, Florida, May 8–11, 2006). The objectives of this study are to 1) describe the prevalence of bacterial vaginosis among U.S. women between the ages of 14 and 49 using nationally representative data from National Health and Nutrition Examination Survey collected from 2001 to 2004, and 2) evaluate whether sociodemographic characteristics, such as race or ethnicity and age, which are correlates in clinical populations, are also associated with the prevalence of bacterial vaginosis in the general population.
MATERIALS AND METHODS
We used data from the National Health and Nutrition Examination Survey (NHANES) samples for the combined intervals 2001–2002 and 2003–2004 to estimate the prevalence of bacterial vaginosis among women in the civilian, noninstitutionalized U.S. population. The National Health and Nutrition Examination Survey, conducted by the National Center for Health Statistics at the Centers for Disease Control and Prevention, was designed to obtain nationally representative information on the health and nutritional status of the population of the United States through interviews and direct physical examinations. Methods describing this national survey have been published elsewhere.11 This study submitted to the Washington University School of Medicine Human Subjects Committee for approval. The study was classified as exempt because it is a population-based study devoid of individual identifiers.
For these analyses, women between the ages of 14 and 49 years with bacterial vaginosis data were included. A total of 747 women were excluded from the analyses because of missing data for bacterial vaginosis. The final sample included data from 3,727 women, which when weighted represents the experience of 65,660,083 U.S. women between the ages of 14 and 49 years.
Female participants in the National Health and Nutrition Examination Survey study between the ages of 14 and 49 years were tested for bacterial vaginosis. Self-collected vaginal swabs were used for the evaluation of bacterial vaginosis. Smears were allowed to air dry before shipment to the processing and analysis laboratory at Magee-Women's Hospital (Pittsburgh, PA). The bacterial vaginosis score for Gram staining was calculated by Nugent's method.12 Scores of 7 or higher were considered positive for bacterial vaginosis, whereas those between 4 and 6 were considered intermediate. Additional details on laboratory procedures have been published.13,14 Prevalence was estimated for the three levels of bacterial vaginosis: positive, negative, and intermediate. For logistic regression analyses, the outcome was defined as bacterial vaginosis confirmed (positive) or not (negative and intermediate).
Self-reported sociodemographic characteristics included in these analyses included age, race or ethnicity, highest level of education received, and poverty/income ratio. Age was categorized by decade (14–19, 20–29, 30–39, and 40–49). Race or ethnicity was categorized as white, non-Hispanic; black, non-Hispanic, Mexican American; or other race/Hispanic. Three levels of education were evaluated: less than high school, completed high school (or general equivalency diploma), or more than high school. Poverty/income ratio was the ratio of the individual's family household income to the federal poverty level and was categorized as less than the federal poverty level (poverty/income ratio less than 1), at or above the federal poverty level (poverty/income ratio 1 to less than 2), or twice the federal poverty level (poverty/income ratio 2 or more).
In this analysis we examined a number of reproductive history variables. Age at first menstruation was categorized as 7–11 years, 12–14 years, 15 years or later, or unknown. Douching within the past 6 months was defined as yes, no, or unknown.
Women between the ages of 14 and 49 years also completed a sexual history questionnaire that included questions on history of sexual intercourse and lifetime and recent number of sexual partners. Because of confidentiality concerns, sexual history data for individuals between the ages of 14 and 19 years are only available at the National Center for Health Statistics Research Data Center. Therefore, all analyses including sexual history were limited to women between the ages of 20 and 49 years of age.
Categorical data were compared by using χ2 tests. Crude and adjusted odds ratios were estimated by using logistic regression. Statistical analyses were conducted with Stata 9.2 (StataCorp, College Station, TX). Specifically, using the svyset command in Stata, we specified the individual weight, primary sampling unit, and stratum. The 2-year individual weights estimated by the National Center of Health Statistics and made available as part of the National Health and Nutrition Examination Survey data set are adjusted to the entire U.S. population based on 2000 Census information. To accommodate the joining of 2001–2002 and 2003–2004 data sets, each adjusted to the U.S. population, the weight for each individual was divided by 2 to provide a single estimate for the entire U.S. population.
The overall and subgroup prevalences of bacterial vaginosis are presented in Table 1. The prevalence of bacterial vaginosis in the general population of the United States was high—almost one woman in three was positive for bacterial vaginosis (29.2%). Although younger women, those between the ages of 14 and 19 years, had a somewhat lower prevalence (23.3%) of bacterial vaginosis, among the 20 years and older group the prevalence was between 28% and 31%. The prevalence of bacterial vaginosis varied significantly with race or ethnicity, education, and poverty/income ratio. Bacterial vaginosis was more common among black, non-Hispanic (51.6%) and Mexican-American (32.1%) women than among white, non-Hispanic women (23.2%). Women with more than a high school education were less likely to be positive for bacterial vaginosis than those with a high school education or less (26% versus 33–34%). Similarly, the prevalence of bacterial vaginosis was lower among those living well above the federal poverty level (24%) compared with those living at or near (34%) or below (37%) the federal poverty level.
Selected reproductive history characteristics were also associated with the prevalence of bacterial vaginosis. Although there was no noticeable trend in the relationship to bacterial vaginosis and age at menarche, women who reported douching in the past 6 months had significantly higher prevalence of bacterial vaginosis than those who did not (45% versus 24%; P<.001).
We examined the association between sexual history and bacterial vaginosis prevalence among women between the ages of 20 and 49 years and found that a history of sexual intercourse and number of male sexual partners were associated with bacterial vaginosis. Women who reported no history of sexual intercourse had rates of bacterial vaginosis that were half that of women who reported a history of intercourse (15% versus 30%). The prevalence of bacterial vaginosis was highest among those who reported the youngest ages of first intercourse: 38% among those who reported first intercourse between the ages of 14 and 19 compared with 30% for those aged 15–19 years and 24% for those whose first intercourse occurred at age 20 or older. Number of male sexual partners during one's lifetime and in the past year was also associated with prevalence of bacterial vaginosis. Prevalence was lowest among women with the fewest male sexual partners in their lifetimes and among those with a single male sexual partner in the past year.
Results from crude and adjusted logistic regression analyses are presented in Table 2. Although the odds of bacterial vaginosis were lower among women between the ages of 14 and 19 years compared with 20–29 year olds, there were no significant differences between 20–29 year olds and those at ages 30–39 years or 40–49 years. Differences observed in the crude prevalence of bacterial vaginosis by race or ethnicity, education, and poverty/income ratio persisted in the adjusted analyses. Black, non-Hispanic women had an odds of bacterial vaginosis three times that of non-Hispanic, white women after adjusting for age, education, and poverty. Similarly, the odds for Mexican-American women were slightly higher when compared with white women. The adjusted analyses confirmed associations of bacterial vaginosis with lower levels of education and living near or below the federal poverty level. These differences remained with the addition of history of douching to the model. The association with black race was attenuated somewhat (OR 2.66, 95% CI 2.18–3.25).
In the subsample of women between the ages of 20 and 49 years, we also examined the association with specific sexual characteristics after adjustment for age, race or ethnicity, education, and poverty level (data not shown). After adjustment for age, race or ethnicity, education, and poverty level, age at first intercourse and number of lifetime male sexual partners were associated with increased odds of bacterial vaginosis. Women who reported later ages of intercourse had somewhat, but not significantly, lower odds of bacterial vaginosis. Women with two to five lifetime male sexual partners did not have a higher odds of bacterial vaginosis (OR 1.03, 95% CI 0.69–1.54) than women who had one lifetime male sexual partner, but women with six or more partners did have a higher odds ratio (OR [6–10 partners] 1.47, 95% CI 1.03–2.09; OR [11 or more partners] 1.62, 95% CI 1.09–2.39). Further, having had two or more male sexual partners in the past year was associated with an increased odds of bacterial vaginosis compared with women with a single male sexual partner (OR 1.46, 95% CI 1.08–1.98).
Although preliminary, these data suggest possible interaction between race or ethnicity and other sociodemographic characteristics. Figure 1 presents the prevalence of bacterial vaginosis by age, education, and poverty/income ratio for the different strata of race or ethnicity. We sought to understand whether the relationship between age, education, and poverty were uniform across race or ethnic groups. For example, an inspection of the association with age in the different groups indicates that bacterial vaginosis prevalence increases with age among white, non-Hispanic women (P=.33, χ2 test), is flat among Mexican-American women (P=.70), and appears to plateau among black, non-Hispanic women from 20 years of age and older (P=.21). Similarly, when examining the association between race or ethnicity and poverty level, the prevalence of bacterial vaginosis decreases slightly, but not significantly, within increasing income for white and Mexican-American women (P=.15 and P=.16, respectively). Among black women, there was little difference in the prevalence of bacterial vaginosis among those living below (55%) or near (57%) the poverty level and a significant decrease in prevalence among those living at two times the poverty level (44%; P=.01). An additional analysis, which evaluated for interactions between race or ethnicity and sociodemographic characteristics, largely confirmed these findings. The findings were suggestive of interactions with age (P=.04), level of education (P=.07), and poverty/income ratio (P=.11) and white, non-Hispanic race or ethnicity. Further, there was a significant interaction between age and Mexican-American ethnicity. There were no significant interactions with black, non-Hispanic race or ethnicity.
Although many studies have examined the prevalence of bacterial vaginosis among different subpopulations, the National Health and Nutrition Examination Survey provides an estimate of the prevalence of bacterial vaginosis in the general U.S. population. The prevalence of bacterial vaginosis in multiple studies of pregnant women ranged from 12% to 21%,5,15,16 was as high as 30% in people seeking health care or seeking termination of pregnancy,17,18 and was over 50% in a population of injection drug users.19 Yen and colleagues3 examined women entering the military and found that 28% of the sexually experienced and 18% of non–sexually experienced women had bacterial vaginosis. Almost one third of all women tested positive for bacterial vaginosis.
Consistent with the existing literature,18,20,21 bacterial vaginosis was more common among black and Mexican-American women. A study in women presenting at a county health center in Michigan found prevalence rates of 42% among black women, 35% among Hispanic women, and 25% among white women.18 The prevalences for white (23%) and Hispanic (32%) women were roughly consistent with those in the general population but somewhat lower than that observed among black women (52%).
Of note, these data support the conclusion that bacterial vaginosis is not exclusively a sexually transmitted condition. Almost 15% of women who reported no history of sexual intercourse had bacterial vaginosis. This finding confirms a previous study by Yen and colleagues,3 in which approximately 18% of sexually inexperienced women were found to be positive for bacterial vaginosis.
Previous studies have found that bacterial vaginosis prevalence increased with age. In a population of individuals seeking STD treatment, 23% of women aged 14–24 years had bacterial vaginosis compared with 33% of women aged 25 years and older.17 Although these national data confirmed that the prevalence of bacterial vaginosis is lower among women 14 years of age, there is no evidence to support the assertion that bacterial vaginosis is associated with age among older women. This contradiction may be a consequence of clinical study designs that disproportionately enroll younger (and higher risk) women.
Socioeconomic status and poverty are also associated with the distribution of bacterial vaginosis in the population. Factors such as Medicaid status, low levels of education, absence of a telephone in the home, occupation, and employment status have all been found to be associated with higher frequency of bacterial vaginosis.18,20,22,23
As found in several other studies,20,22,24,25 douching was associated with higher prevalence of bacterial vaginosis. Further many studies have found that the prevalence of douching is higher among African-American populations. In this sample 15–17% of white and Mexican-American women reported douching in the past 6 months compared with 44% of black women. The addition of douching to the adjusted model resulted in a change of the effect size of black race, but both factors remained significant predictors of bacterial vaginosis independently. Because this analysis is cross-sectional, it is not possible to ascertain whether the association with douching is causal or a result of attempts to self-treat vaginal symptoms.
Bacterial vaginosis is common among the general population of women in the United States. In fact, the prevalence of women with bacterial vaginosis in National Health and Nutrition Examination Survey was comparable with that in many treatment-seeking populations. These data confirm what has been learned about the sociodemographic distribution of bacterial vaginosis from clinical populations, namely, that race or ethnicity, education, and poverty are all associated with bacterial vaginosis prevalence. One contradiction, however, is the association with age. In contrast with previous studies, there is not an increasing prevalence with increasing age. Further, these findings indicate that the relationship between demographic characteristics and bacterial vaginosis may vary by race or ethnicity. It was not clear from subgroup analyses that associations with demographic characteristics were consistent across race or ethnicity. Bacterial vaginosis is an important predictor of adverse reproductive outcomes, and more complete understanding of the dynamics connecting these sociodemographic characteristics will allow for the creation of targeted interventions. Additional waves of National Health and Nutrition Examination Survey data will be helpful in evaluating this question in detail in the future.
1. Ness RB, Kip KE, Hillier SL, Soper DE, Stamm CA, Sweet RL, et al. A cluster analysis of bacterial vaginosis-associated microflora and pelvic inflammatory disease. Am J Epidemiol 2005;162:585–90.
2. Bradshaw CS, Morton AN, Garland SM, Morris MB, Moss LM, Fairley CK. Higher-risk behavioral practices associated with bacterial vaginosis compared with vaginal candidiasis. Obstet Gynecol 2005;106:105–14.
3. Yen S, Shafer MA, Moncada J, Campbell CJ, Flinn SD, Boyer CB. Bacterial vaginosis in sexually experienced and non-sexually experienced young women entering the military. Obstet Gynecol 2003;102:927–33.
4. Moodley P, Connolly C, Sturm AW. Interrelationships among human immunodeficiency virus type 1 infection, bacterial vaginosis, trichomoniasis, and the presence of yeasts. J Infect Dis 2002;185:69–73.
5. Royce RA, Thorp J, Granados JL, Savitz DA. Bacterial vaginosis associated with HIV infection in pregnant women from North Carolina. J Acquir Immune Defic Syndr Hum Retrovirol 1999;20:382–6.
6. Cohen CR, Duerr A, Pruithithada N, Rugpao S, Hillier S, Garcia P, et al. Bacterial vaginosis and HIV seroprevalence among female commercial sex workers in Chiang Mai, Thailand. AIDS 1995;9:1093–7.
7. Haggerty CL, Hillier SL, Bass DC, Ness RB. Bacterial vaginosis and anaerobic bacteria are associated with endometritis. Clin Infect Dis 2004;39:990–5.
8. Evans BA, Kell PD, Bond RA, MacRae KD, Slomka MJ, Brown DW. Predictors of seropositivity to herpes simplex virus type 2 in women. Int J STD AIDS 2003;14:30–6.
9. Forsum U, Holst E, Larsson PG, Vasquez A, Jakobsson T, Mattsby-Baltzer I. Bacterial vaginosis: a microbiological and immunological enigma. APMIS 2005;113:81–90.
10. Klebanoff MA, Schwebke JR, Zhang J, Nansel TR, Yu KF, Andrews WW. Vulvovaginal symptoms in women with bacterial vaginosis. Obstet Gynecol 2004;104:267–72.
11. Centers for Disease Control and Prevention (CDC). National Center for Health Statistics (NCHS). National Health and Nutrition Examination Analytic and Reporting Guidelines. Hyattsville (MD): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2005.
12. Nugent RP, Krohn MA, Hillier SL. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation. J Clin Microbiol 1991;29:297–301.
13. Laboratory procedure manual: bacterial vaginosis. Hyattsville (MD): National Center for Health Statistics, Centers for Disease Control and Prevention; 2006.
14. MEC Laboratory component: trichomonas vaginalis and bacterial vaginosis. Hyattsville (MD): National Center for Health Statistics, Centers for Disease Control and Prevention; 2006.
15. Hillier SL, Nugent RP, Eschenbach DA, Krohn MA, Gibbs RS, Martin DH, et al. Association between bacterial vaginosis and preterm delivery of a low-birth-weight infant. The Vaginal Infections and Prematurity Study Group. N Engl J Med 1995;333:1737–42.
16. Hay PE, Lamont RF, Taylor-Robinson D, Morgan DJ, Ison C, Pearson J. Abnormal bacterial colonisation of the genital tract and subsequent preterm delivery and late miscarriage. BMJ 1994;308:295–8.
17. Moi H. Prevalence of bacterial vaginosis and its association with genital infections, inflammation, and contraceptive methods in women attending sexually transmitted disease and primary health clinics. Int J STD AIDS 1990;1:86–94.
18. Holzman C, Leventhal JM, Qiu H, Jones NM, Wang J. Factors linked to bacterial vaginosis in nonpregnant women. Am J Public Health 2001;91:1664–70.
19. Plitt SS, Garfein RS, Gaydos CA, Strathdee SA, Sherman SG, Taha TE. Prevalence and correlates of chlamydia trachomatis, neisseria gonorrhoeae, trichomonas vaginalis infections, and bacterial vaginosis among a cohort of young injection drug users in Baltimore, Maryland. Sex Transm Dis 2005;32:446–53.
20. Rajamanoharan S, Low N, Jones SB, Pozniak AL. Bacterial vaginosis, ethnicity, and the use of genital cleaning agents: a case control study. Sex Transm Dis 1999;26:404–9.
21. Ness RB, Hillier S, Richter HE, Soper DE, Stamm C, Bass DC, et al. Can known risk factors explain racial differences in the occurrence of bacterial vaginosis? J Natl Med Assoc 2003;95:201–12.
22. Meis PJ, Goldenberg RL, Mercer BM, Iams JD, Moawad AH, Miodovnik M et al. Preterm Prediction Study: is socioeconomic status a risk factor for bacterial vaginosis in Black or in White women? Am J Perinatol 2000;17:41–5.
23. Hart G. Factors associated with trichomoniasis, candidiasis and bacterial vaginosis. Int J STD AIDS 1993;4:21–5.
24. Ness RB, Hillier SL, Richter HE, Soper DE, Stamm C, McGregor J, et al. Douching in relation to bacterial vaginosis, lactobacilli, and facultative bacteria in the vagina. Obstet Gynecol 2002;100:765.
25. Cottrell BH. Vaginal douching practices of women in eight Florida panhandle counties. J Obstet Gynecol Neonatal Nurs 2006;35:24–33.