Twenty-seven percent of U.S. females aged 14–59 are infected with at least one type of human papillomavirus (HPV); 15% are positive for high-risk types, which can cause cervical, genital or anal cancer, and 18% for low-risk types, according to a survey conducted in 2003–2004.1 Among all age-groups, the highest prevalence of HPV infection—45%—is among 20–24-year-olds. HPV infection is associated with being aged 20–24, being unmarried, having had three or more lifetime sexual partners and having had any partners in the last year.
To provide an estimate of HPV prevalence among females before 2006, when the Food and Drug Administration (FDA) approved an HPV vaccine against four types (two high-risk and two low-risk), researchers analyzed data from 1,921 females who participated in the 2003–2004 National Health and Nutrition Examination Survey, which uses a nationally representative sample of noninstitutionalized U.S. women aged 14–59. Self-collected cervicovaginal samples taken at a mobile examination center were tested for HPV DNA, and data were collected on participants’ demographic characteristics, as well as sexual history. The weighted prevalence of HPV types was compared using the McNemar test, and associations between the presence of any HPV and personal characteristics were assessed using the Wald chi-square test and multivariate logistic regression analysis (which was limited to women aged 18–59).
Twenty-seven percent of surveyed females tested positive for HPV; using 2000 census data, the researchers calculate that this represents nearly 25 million infected individuals. The highest rate of infection was among 20–24-year-olds, 45% of whom were infected; prevalence was 20% among those aged 50–59 and 25–28% among other age-groups. Prevalence increased with each year between the ages of 14 and 24. Among sexually active females, 49% of those aged 20–24 were infected, as were 40% of those aged 14–19; rates were lower for women 25 or older. Overall, 39% of blacks tested positive, as did 24% of whites and Mexican Americans. HPV prevalence differed depending on marital status: 46% among those who lived with a partner; 41% among those who were widowed, divorced or separated; 31% among those who had never married; and 17% among those who were married. Prevalence decreased from 35% among females who had not graduated from high school to 25% among those with any postsecondary education. Thirty-eight percent of those living below the poverty index tested positive, as did 24% of those at or above the index.
Fifteen percent of all surveyed females were infected with a high-risk HPV type, and 18% were infected with a low-risk type; rates of both types were highest among those aged 20–24. Infection with high-risk types decreased after age 29, and infection with low-risk types leveled off after age 39. The HPV types that can be prevented by the FDA-approved vaccine were found in 0.1–1.5% of females; overall, 3.4% tested positive for at least one of these types. Sixty percent of females who had any HPV infection had only one type of the virus, 24% had two and 16% had three or more.
Multivariate logistic regression analysis of data on women aged 18–59 found a number of factors to be independently associated with HPV infection. Women aged 20–24 were more likely than those 30 or older to be infected (odds ratio, 2.2). Marital status was a significant predictor of infection: Compared with married women, those who were formerly married, never-married or living with a partner had higher odds of being infected (3.1, 2.2 and 3.4, respectively). Furthermore, women who had had three or more lifetime sexual partners had higher odds of being infected than those with a single lifetime partner (2.7), and compared with women who had had no partners in the last year, those who had had one, two, or three or more sexual partners in the last year had elevated odds (2.1, 4.0 and 4.1, respectively).
The researchers note several limitations of their study. Women who declined to participate were more likely than respondents to be of "other" race or ethnicity, to be younger than 40, to have been born outside the United States or Mexico, and to have never had sex. Also, cervicovaginal samples may not detect the same HPV types as cervical mucosa samples collected in other studies, and prevalence rates determined by DNA tests underestimate the cumulative incidence of HPV because these tests do not indicate past exposure. Nonetheless, the researchers state that this is the first study to provide a national estimate over a broad age range, and they believe that its findings will be useful in assessing "the wide-scale impact of the vaccine for reducing infection and could help guide models evaluating impact and cost-effectiveness." —J. Thomas
REFERENCE
1. Dunne EF et al., Prevalence of HPV infection among females in the United States, Journal of the American Medical Association, 2007, 297(8):813–819.