Socio-demographic and behavioral correlates of herpes simplex virus type 1 and 2 infections and co-infections among adults in the USA
Summary
Objectives
Although herpes simplex virus (HSV)-1 and HSV-2 may co-exist and interact, some epidemiologic features including geographical distribution, secular trends, route of transmission, and established risk factors may distinguish these HSV sub-types. With recent data indicating a link between genital herpes and either strain, a re-evaluation of risk factors for HSV-1/HSV-2 infection and co-infection is needed.
Methods
We used the 1999–2004 National Health and Nutrition Examination Study (NHANES) data and logistic regression modeling to identify socio-demographic and behavioral risk factors that can independently predict HSV-1/HSV-2 infection and co-infection.
Results
Approximately 48% were positive for HSV-1 alone, 7% were positive for HSV-2 alone and 12% were co-infected with HSV-1 and HSV-2. History of genital herpes was found in individuals infected with either or both HSV sub-types. Whereas age, sex, race, and level of education independently predicted all three outcomes, lifetime sexual activity as well as use of tobacco products and recreational drugs mainly correlated with HSV-2 infection and HSV-1/HSV-2 co-infection. Lifetime use of alcohol was not consistently associated with HSV-1/HSV-2 infection and co-infection.
Conclusions
Sexual activity played an important role for HSV-2 and was potentially important for HSV-1, with implications for healthcare practice and vaccine development.
Introduction
Nearly eight distinct herpesviruses are known to cause infectious diseases in human populations.1,2 This Herpesviridae group includes herpes simplex viruses (HSV), varicella zoster virus, cytomegalovirus, Epstein–Barr virus, and Kaposi’s sarcoma-associated viruses.1,3 Believed to be prevalent in 60–95% of human adults, HSV infection is frequently asymptomatic and exists as two main sub-types, HSV-1 and HSV-2, both of which have been linked with significant disability and healthcare costs.1,4 HSV infection can also be fatal, especially in neonates and immune suppressed persons.1,4 Once acquired, HSV infections are lifelong non-curable conditions, with periods of latency and recurrence.1,4 Spread of an HSV infection occurs as a result of symptomatic or asymptomatic reactivation, which can be either spontaneous or triggered by external stimuli.1,4 In the absence of effective vaccination, various treatments are currently used to alleviate symptoms and reduce the likelihood of horizontal and vertical transmission.1 Because it is not a reportable disease, reliable data on HSV prevalence is inconsistently available worldwide. Yet, it has been established that patterns of HSV-1/HSV-2 infection and co-infection vary by geographic region and by ‘high-risk’ status for acquiring sexually transmitted diseases (STDs).5,6 In the USA, it is estimated that 40–63% of people are seropositive for HSV-1, whereas 16–18% are seropositive for HSV-2,1,2 ranking above European countries and below sub-Saharan Africa.4,7 Since high-risk groups for either or both HSV sub-types need to be identified prior to public health intervention, the epidemiology of co-infection with HSV-1 and HSV-2 is of particular interest. Whether or not infection with one HSV type can protect against the other or reduce clinical manifestations of the other remains subject to debate.1,4 Overall evidence suggests that antibodies against one type of HSV (e.g., HSV-1) provide cross-immunity, ameliorating the severity, duration, and frequency of a subsequent infection by the other serotype (e.g., HSV-2).1
Despite the fact that HSV-1 and HSV-2 may co-exist and interact, many epidemiologic features distinguish these two viral sub-types. First, HSV-1 infection has been primarily linked with non-genital diseases.1 By contrast, HSV-2 is thought to be the agent responsible for genital herpes.1 Greater attention has been paid to HSV-2 than to HSV-1 due to its putative role in facilitating the acquisition, progression and spread of HIV.1,4,8–11 Recent information indicates that both HSV-1 and HSV-2 can be linked to genital herpes.1,4 In fact, an increasing number of genital herpes cases are being attributed to HSV-1 rather than HSV-2 exposure.4 Diminished exposure to HSV-1 during childhood and risky sexual behaviors that promote the transmission of HSV-1 through the oral–genital route may be responsible for this finding.1,4 Second, secular trends indicate that HSV-2 seroprevalence is on the rise in many developed and developing countries, whereas HSV-1 seroprevalence has not changed substantially over time.1,4 Third, HSV-1 and HSV-2 share some of their established risk factors. HSV-1 seroprevalence is influenced by age, race, geographical region, and socioeconomic status, whereas HSV-2 seroprevalence is affected by age, sex, race, socioeconomic status, STD history, age at first intercourse, and number of lifetime sexual partners.1
Due to recent changes in the epidemiology of HSV-1 and HSV-2, a renewed investigation of personal characteristics that may distinguish HSV-1 and HSV-2 infections and co-infections is warranted. Both modifiable and non-modifiable risk factors need to be examined to inform policy-makers in efforts to reduce the burden of HSV infection and its associated health problems. In the current study we used a nationally representative sample to identify socio-demographic and behavioral risk factors associated with HSV-1 infection alone, HSV-2 infection alone, and co-infection with HSV-1 and HSV-2 among adults aged 18–49 years in the USA. We specifically hypothesized that risk factors for one HSV strain will differ among those who are co-infected with the other HSV stain compared to those who are not.
Methods
Data source
We analyzed existing data from the National Health and Nutrition Examination Study (NHANES). Conducted by the National Center for Health Statistics (NCHS) and the Centers for Disease Control and Prevention (CDC), NHANES is a series of cross-sectional surveys originally designed for the evaluation of health and nutritional characteristics of the civilian population of the USA. All NHANES surveys use a stratified, multistage probability cluster sampling design. Demographic, socioeconomic, and health data were collected by trained staff using household interviews. In addition, a mobile examination center run by health professionals collected numerous measurements including anthropometric, blood pressure, and laboratory tests, either on all or a sub-group of study participants. Detailed descriptions of the NHANES study design and data collection procedures have been published elsewhere.12,13 Since 1999, NHANES has become a continuous surveillance system. We therefore combined the 1999–2000, 2001–2002, and 2003–2004 NHANES datasets for the current analyses. The study was approved by the Eastern Virginia Medical School Institutional Review Board.
Measures – HSV-1 and HSV-2
The 1999–2004 NHANES surveys include laboratory data for several STDs. The current analyses were focused on HSV-1 and HSV-2 infection and co-infection among adults aged 18–49 years. In NHANES, sera were tested for antibodies against HSV-1 and HSV-2.5 Monoclonal antibodies and affinity chromatography were used to purify viral glycoproteins specific to HSV-1 (designated gG-1) and HSV-2 (designated gG-2) and provide antigens for type-specific herpes serology assays.5 Antibodies reactive to those antigens were detected using a highly sensitive test capable of discriminating between HSV-1 and HSV-2,5 namely the solid-phase enzymatic immunodot assay.14 One caveat to the immunoassay is that it lacks sensitivity for early HSV-1 infection.14 HSV-2 seropositivity was further confirmed by a gG-2 monoclonal antibody inhibition assay.5 The primary outcome of interest was HSV-1/HSV-2 infection and co-infection defined as ‘HSV-1 alone’, ‘HSV-2 alone’, ‘both HSV-1 and HSV-2’ or ‘neither HSV-1 nor HSV-2’; the latter group was selected as reference. Self-reported history of genital herpes and other STDs was also evaluated through interview data. It is worth noting that no enquiries were made with regard to presence of vesicles, ulcers, or non-specific symptoms such as itching or burning. Rather, participants were asked ‘Has a doctor or other healthcare professional ever told you that you had genital herpes?’
Socio-demographic and risk behavioral factors
Data on demographic, socioeconomic, and tobacco use characteristics were ascertained for all survey participants through computer-assisted personal interviews (CAPI).15 Sexual behaviors and alcohol and recreational drug use were evaluated in different age groups applying distinct methodologies. For instance, alcohol use was assessed for individuals 20 years and older through CAPI.15 Recreational drug use and sexual behaviors were assessed using an audio computer-assisted personal self-interview (ACASI) questionnaire for individuals in the age range of 12–59 years and 14–59 years, respectively.16–18
Statistical analysis
Descriptive, bivariate, and multivariate analyses were conducted using STATA version 8. Two-sided statistical tests were performed at an alpha level of 0.05. Using survey commands, we applied recommended sub-sample weights for the period of 1999–2004. Mobile examination center (MEC) exam weights were used for all analyses. Masked variance units were used to estimate variances through a Taylor series linearization method. Crude and adjusted odds ratios (COR and AOR) and their 95% confidence intervals (CI) were computed, taking sampling weights into consideration. These weights were defined to represent the US civilian, non-institutionalized population while accounting for over-sampling of certain age and ethnic groups and interview non-response.2 Bivariate associations were analyzed using Pearson’s Chi-square test for independence. Initially, we evaluated the relationship of HSV-1/HSV-2 infection and co-infection with selected socio-demographic characteristics, namely sex, age group, race, and education. Then, we examined whether lifetime use of tobacco products (cigarettes, cigars, pipes, snuff, chew), alcohol, and recreational drugs, as well as sexual activity, correlated with HSV-1/HSV-2 infection and co-infection, after adjustment for age and sex. Sexual activity was defined in terms of lifetime experience with and age at initiation of vaginal, oral, or anal sex; data on specific sexual practices such as oral sex or number of simultaneous sexual partners were not available. Finally, a multinomial logistic regression model for correlates of HSV-1/HSV-2 infection and co-infection was created. The model included socio-demographic (age, sex, race, education) and behavioral (lifetime experiences with tobacco, alcohol, recreational drugs and sexual intercourse) characteristics that were plausibly linked to either or both HSV strains. In this model, the dependent variable was defined as four categories, namely ‘non-infected (reference group)’, ‘HSV-1 infected alone’, ‘HSV-2 infected alone’ and ‘HSV-1 and HSV-2 co-infected’; two-way interaction terms were evaluated for statistical significance within the multinomial model.
Results
HSV-1/HSV-2 infection and co-infection
In the 1999–2004 NHANES, laboratory tests used to measure HSV-1 and HSV-2 seroprevalence were performed on a total of 8277 survey participants aged 18–49 years. Furthermore, 5850 subjects had interview data on past diagnosis with genital herpes. Unweighted analyses suggested the following: of 204 participants with self-reported genital herpes, 21.6% tested positive for HSV-1 alone, 34.8% tested positive for HSV-2 alone, and 39.2% were co-infected with HSV-1 and HSV-2; Of 5646 participants with no self-reported genital herpes, 52.9% tested positive for HSV-1 alone, 6.4% tested positive for HSV-2 alone, and 14.6% were co-infected with HSV-1 and HSV-2. Further analyses indicated that the mean age of participants who tested positive for HSV-1 alone and had a history of genital herpes was 35.9 (± 6.9) years; their mean age at initiation of sexual activity was 16.6 (± 2.4) years. Neither their current age nor their age at initiation of sexual activity was significantly different from that of participants who tested positive for HSV-2 alone and reported a history of genital herpes; however, the distribution by race was significantly different among these two groups (p < 0.0001), with a greater percentage of non-Hispanic whites in the former than in the latter group. Taking sampling weights into consideration, it was estimated that 48.4% of the study population was positive for HSV-1 alone, 7.4% was positive for HSV-2 alone, and 11.8% was co-infected with HSV-1 and HSV-2; the remaining 32.4% was infected with neither HSV-1 nor HSV-2, and chosen as the reference group. Figure 1 presents the percentage of seronegative individuals according to sex and age group; this subset is susceptible to either HSV strain and would therefore benefit from vaccine development. HSV-1/HSV-2 susceptibility appeared to decline steadily between 18 and 39 years of age (ptrend < 0.0001); furthermore, males were consistently more susceptible to HSV-1/HSV-2 acquisition than females regardless of age (p < 0.0001).
Socio-demographic factors
Prevalence rates of HSV-1/HSV-2 infection and co-infection by socio-demographic characteristics are presented in Table 1. Females had more than twice the odds of being infected with HSV-2 alone (COR 2.4, 95% CI 1.9, 2.9) or of being co-infected with HSV-1 and HSV-2 (COR 2.8, 95% CI 2.4, 3.3) compared to males. A consistently increasing trend by age was noted for HSV-1/HSV-2 co-infection. By contrast, the steady increase in HSV-2 infection alone was noted until the age of 40 years and no clear trend by age was noted for HSV-1 infection alone. Compared to non-Hispanic whites, non-Hispanic blacks were at elevated odds of being infected with HSV-1 alone, HSV-2 alone, or of being co-infected with HSV-1 and HSV-2. By contrast, Hispanics were more likely than non-Hispanic whites to be infected with HSV-1 alone or to be co-infected with HSV-1 and HSV-2. Individuals who achieved high school education or better were less likely than those who achieved less than a high school education to be infected with HSV-1 alone, HSV-2 alone, or both.
Table 1
Infection and co-infection with herpes simplex virus-1 and -2 by socio-demographic characteristics, USA, 1999–2004
| Total | Not infected with HSV-1 or HSV-2 | HSV-1 alone | HSV-2 alone | HSV-1 and HSV-2 | ||||
|---|---|---|---|---|---|---|---|---|
| na | nb | COR (95% CI) | nc | COR (95% CI) | nd | COR (95% CI) | ||
| Sex | ||||||||
| Male | 3852 | 1239 | 2087 | 1.0 | 176 | 1.0 | 350 | 1.0 |
| Female | 4425 | 1130 | 2196 | 1.1 (1.0, 1.3) | 349 | 2.4 (1.9, 2.9) | 750 | 2.8 (2.4, 3.3) |
| Age group (years) | ||||||||
| <20 | 1472 | 664 | 745 | 1.0 | 34 | 1.0 | 29 | 1.0 |
| 20–24 | 1282 | 448 | 694 | 1.9 (1.5, 2.3) | 62 | 3.1 (1.9, 4.8) | 78 | 4.3 (2.8, 6.7) |
| 25–29 | 1137 | 366 | 585 | 1.9 (1.5, 2.3) | 75 | 4.8 (2.9, 7.9) | 111 | 9.5 (6.0, 14.9) |
| 30–34 | 1138 | 289 | 599 | 2.4 (1.9, 3.1) | 86 | 7.7 (4.9, 12.1) | 164 | 17.9 (11.0, 29.4) |
| 35–39 | 1089 | 221 | 573 | 3.3 (2.5, 4.2) | 90 | 11.5 (6.9, 19.3) | 205 | 32.5 (21.0, 50.4) |
| 40–44 | 1157 | 219 | 595 | 2.8 (2.1, 3.6) | 92 | 9.4 (6.4, 13.8) | 251 | 28.9 (18.9, 44.2) |
| 45–49 | 1002 | 170 | 483 | 3.2 (2.4, 4.3) | 86 | 13.3 (8.4, 21.2) | 263 | 41.3 (25.9, 65.8) |
| Race | ||||||||
| Non-Hispanic white | 3457 | 1467 | 1524 | 1.0 | 230 | 1.0 | 236 | 1.0 |
| Non-Hispanic black | 1878 | 365 | 769 | 1.9 (1.6, 2.4) | 231 | 4.6 (3.5, 5.9) | 513 | 10.4 (7.7, 14.0) |
| Hispanic | 2277 | 384 | 1611 | 4.3 (3.4, 5.5) | 38 | 0.8 (0.5, 1.3) | 244 | 4.0 (2.9, 5.5) |
| Other race | 665 | 161 | 370 | 2.2 (1.7, 2.8) | 26 | 0.9 (0.6, 1.4) | 108 | 3.7 (2.3, 5.8) |
| Education | 0 | |||||||
| Less than high school | 2504 | 452 | 1507 | 1.0 | 113 | 1.0 | 432 | 1.0 |
| High school | 2145 | 631 | 1125 | 0.5 (0.4, 0.7) | 121 | 0.7 (0.5, 1.1) | 268 | 0.3 (0.3, 0.5) |
| Greater than high school | 3622 | 1294 | 1637 | 0.4 (0.3, 0.5) | 290 | 0.7 (0.5, 1.0) | 401 | 0.2 (0.2, 0.3) |
COR, crude odds ratios (estimated taking sampling design into consideration); CI, confidence intervals (estimated taking sampling design into consideration).
Risk behavioral factors
Table 2 presents associations of HSV-1/HSV-2 infection and co-infection with lifetime use of tobacco, alcohol, and recreational drugs, as well as sexual activity, after controlling for age and sex. Having ever used any tobacco product, defined as cigarette, cigar, pipe, snuff or chewable tobacco, was positively correlated with all three outcomes. Participants who had ever smoked tobacco were more likely to be infected with HSV-1 alone (AOR 1.2, 95% CI 1.0, 1.4), HSV-2 alone (AOR 1.9, 95% CI 1.4, 2.6), or both (AOR 2.0, 95% CI 1.6, 2.6). Lifetime use of alcohol was associated with a higher chance of HSV-2 infection alone (AOR 1.8, 95% CI 1.0, 3.2), but was unrelated to HSV-1 infection alone or HSV-1/HSV-2 co-infection. Similarly, the odds of infection with HSV-2 alone and co-infection with HSV-1 and HSV-2 were nearly twice as high among users versus non-users of recreational drugs. Compared to sexually non-active individuals, the likelihoods of HSV-1 infection alone, HSV-2 infection alone, and HSV-1/HSV-2 co-infection were consistently increased for those who initiated sexual activity prior to 13 years of age.
Table 2
Infection and co-infection with herpes simplex virus-1 and -2 by behavioral characteristics, USA, 1999–2004
| Total | Not infected with HSV-1 or HSV-2 | HSV-1 alone | HSV-2 alone | HSV-1 and HSV-2 | ||||
|---|---|---|---|---|---|---|---|---|
| na | nb | AOR (95% CI) | nc | AOR (95% CI) | nd | AOR (95% CI) | ||
| Tobacco use | ||||||||
| Never | 3506 | 935 | 1882 | 1.0 | 221 | 1.0 | 468 | 1.0 |
| Ever | 3258 | 763 | 1624 | 1.2 (1.0, 1.4) | 269 | 1.9 (1.4, 2.6) | 602 | 2.0 (1.6, 2.6) |
| Alcohol use | ||||||||
| Never | 810 | 182 | 467 | 1.0 | 41 | 1.0 | 120 | 1.0 |
| Ever | 5414 | 1407 | 2772 | 0.8 (0.6, 1.0) | 405 | 1.8 (1.0, 3.2) | 830 | 1.1 (0.7, 1.8) |
| Recreational drug use | ||||||||
| Never | 4807 | 1253 | 2602 | 1.0 | 295 | 1.0 | 657 | 1.0 |
| Ever | 1340 | 315 | 601 | 0.9 (0.8, 1.1) | 143 | 2.5 (1.8, 3.3) | 281 | 1.9 (1.5, 2.5) |
| Sexual activity | ||||||||
| Never | 288 | 87 | 165 | 1.0 | 5 | 1.0 | 31 | 1.0 |
| Initiated <13 years | 317 | 52 | 146 | 1.5 (0.9, 2.3) | 30 | 4.5 (1.9, 10.3) | 89 | 3.3 (1.5, 7.1) |
| Initiated ≥13 years | 5542 | 1431 | 2892 | 1.1 (0.8, 1.4) | 404 | 1.9 (0.9, 3.8) | 815 | 1.2 (0.6, 2.3) |
Note: Odds ratio for each behavioral characteristic was adjusted for the socio-demographic variables sex and age (years).
AOR, adjusted odds ratios (estimated taking sampling design into consideration); CI, confidence intervals (estimated taking sampling design into consideration).
Risk factors for HSV-1/HSV-2 infection and co-infection
A multinomial logistic model for correlates of HSV-1/HSV-2 infection and co-infection is displayed in Table 3. None of the two-way interactions examined in the multinomial model were found to be statistically significant. After adjustment, age, female sex, black non-Hispanic race, level of education, and having ever used tobacco products were clearly important predictors of HSV-1 infection alone, HSV-2 infection alone, and HSV-1/HSV-2 co-infection. While alcohol use indicators were not significantly related to the outcomes of interest, HSV-2 infection alone and HSV-1/HSV-2 co-infection were nearly twice as high among users of recreational drugs. Whereas sexually active individuals were substantially more likely than their counterparts to be infected with HSV-2 alone, lifetime experience with vaginal, oral, or anal sex was also associated with a significantly increased odds of HSV-1 infection alone and HSV-1/HSV-2 co-infection. Broadly speaking, risk factors for HSV-1 infection were found to be different among individuals who were co-infected with HSV-2 compared to those who were not; by contrast, risk factors for HSV-2 infection were mostly similar among individuals who were co-infected with HSV-1 compared to those who were not.
Table 3
Multinomial logistic model for socio-demographic and behavioral correlates of infection and co-infection with herpes simplex virus type-1 and -2, USA, 1999–2004
| HSV-1 alonea | HSV-2 alonea | HSV-1 and HSV-2a | |
|---|---|---|---|
| AOR (95% CI) | AOR (95% CI) | AOR (95% CI) | |
| Sex | |||
| Male | 1.0 | 1.0 | 1.0 |
| Female | 1.3 (1.1, 1.4) | 3.0 (2.3, 4.0) | 4.6 (3.6, 5.9) |
| Age (years) | 1.0 (1.0, 1.0) | 1.7 (1.0, 1.1) | 1.1 (1.1, 1.1) |
| Race | |||
| Non-Hispanic white | 1.0 | 1.0 | 1.0 |
| Non-Hispanic black | 1.9 (1.5, 2.5) | 6.2 (4.5, 8.4) | 14.3 (9.9, 20.7) |
| Hispanic | 4.2 (3.2, 5.4) | 1.3 (0.8, 2.2) | 4.7 (3.2, 6.9) |
| Other race | 2.8 (2.1, 3.7) | 1.6 (1.0, 2.6) | 5.9 (3.6, 9.9) |
| Education | |||
| Less than high school | 1.0 | 1.0 | 1.0 |
| High school | 0.6 (0.5, 0.8) | 0.6 (0.3, 0.9) | 0.3 (0.2, 0.5) |
| More than high school | 0.4 (0.3, 0.5) | 0.6 (0.4, 0.9) | 0.2 (0.2, 0.3) |
| Tobacco use | |||
| Ever | 1.0 | 1.0 | 1.0 |
| Never | 1.2 (1.0, 1.5) | 1.5 (1.0, 2.2) | 2.1 (1.5, 2.8) |
| Alcohol use | |||
| Ever | 0.8 (0.6, 0.9) | 1.2 (0.7, 2.1) | 1.0 (0.7, 1.5) |
| Never | 1.0 | 1.0 | 1.0 |
| Recreational drug use | |||
| Ever | 0.9 (0.8, 1.1) | 2.2 (1.6, 3.1) | 1.9 (1.4, 2.5) |
| Never | 1.0 | 1.0 | 1.0 |
| Sexual activity | |||
| Initiated <13 years | 1.9 (1.0, 3.3) | 7.3 (1.7, 31.3) | 3.4 (1.6, 7.4) |
| Initiated ≥ 13 years | 1.7 (1.1, 2.6) | 6.8 (1.9, 24.1) | 2.2 (1.3, 3.7) |
| Never | 1.0 | 1.0 | 1.0 |
Note: Odds ratio for each socio-demographic or behavioral characteristic was adjusted for all other variables in the model.
AOR, adjusted odds ratio; CI, confidence intervals.
Discussion
Consistent with past research, we estimated that the prevalence of HSV-1 was considerably higher than that of HSV-2. Our results indicate that HSV-1 infection is prevalent in 60% while HSV-2 infection is prevalent in 19% of NHANES adults 18–49 years of age. These rates are of the same order of magnitude as those reported in other industrialized countries. In a recent study, Pebody et al. described the seroepidemiology of HSV-1 and HSV-2 in eight European countries.19 Age-standardized HSV-1 seroprevalence ranged from 52% in Finland to 84% in Bulgaria. Among individuals 12 years and older, age-standardized HSV-2 seroprevalence ranged from 4% in England and Wales to 24% in Bulgaria.19 Our study is among few to have used nationally representative data in an attempt to identify high-risk groups for HSV-1/HSV-2 infection and co-infection2,21 and the first to construct a multinomial logistic model for socio-demographic and behavioral correlates of infection with HSV-1 alone, HSV-2 alone, and HSV-1/HSV-2 co-infection.
In a systematic review, evidence for HSV-1 and HSV-2 as causative agents of oro-labial and genital diseases was examined.4 While HSV-1 was mostly correlated with oro-labial disease acquired during childhood, genital disease occurred following sexual debut and was mainly attributed to HSV-2.4 Nevertheless, HSV-2 was still capable of causing oro-labial herpes and HSV-1 of causing genital herpes.4 The authors suggested that reduced childhood exposure and immunity to HSV-1 and increased oro-genital contact may have resulted in more genital herpes being associated with HSV-1.4 Although the specific etiology cannot be determined, our analyses indicate that either HSV-1 or HSV-2 can be the agent responsible for genital herpes. Of 204 participants with a history of genital herpes, 21.6% tested positive for HSV-1 alone, 34.8% tested positive for HSV-2 alone, and 39.2% were co-infected with HSV-1 and HSV-2. The symptoms for genital herpes are non-specific; it is, therefore, possible that individuals who were seronegative for HSV-2 and seropositive for HSV-1 reported genitourinary disease associated with HSV-1 although alternative causes such as other STDs and early HSV-2 infection could not be excluded. Since HSV-1 prevalence was almost three times higher than that of HSV-2, it is not surprising that a relatively large proportion (11.8%) was co-infected with HSV-1 and HSV-2. However, given the cross-sectional design, it is not possible to ascertain which of these viral sub-types was first to invade the host or cause genital herpes. Since such infections often go unrecognized, incident cases of HSV-1 or HSV-2 are not easily detected. Therefore, socio-demographic and behavioral characteristics associated with HSV-1, HSV-2, or both infections are merely correlates rather than true risk factors. Nevertheless, it is worth noting that HSV-1 was often the only etiologic agent for self-reported genital herpes. Specifically, 21.6% of genital herpes was associated with HSV-1 alone. Because HSV-2-associated genital ulceration or dense infiltration of immune cells into the genital tract during symptomatic and asymptomatic shedding can facilitate the acquisition, transmission, and progression of HIV, further studies are needed to evaluate whether the same relationship exists between genital HSV-1 and HIV infections. In the meantime, health education and preventive efforts through application of screening tests should be aimed at both HSV sub-types. On the other hand, individuals who were co-infected with HSV-1 and HSV-2 had similar overall characteristics to individuals infected with only HSV-2. Thus, risky sexual behaviors may be the driving force behind HSV-2 infection and HSV-1/HSV-2 co-infection.
In a previous study, Xu et al.2 used 1988–1994 and 1999–2004 NHANES data to assess secular trends in HSV-1 and HSV-2 in the population of the USA. Xu et al.2 indicated that despite recent declines in HSV-1/HSV-2 prevalence rates, HSV-1 (but not HSV-2) seropositive individuals had a greater proportion of self-reported genital herpes in 1999–2004 than in 1988–1994 (1.8% vs. 0.4%; p < 0.001). These results imply that the incidence of genital herpes caused by HSV-1 may be increasing.2 Although genital herpes has been associated with either HSV-1 or HSV-2, few studies have simultaneously analyzed correlates of these two viral sub-types. Most of the current evidence comes from ‘high-risk’ populations, including pregnant and non-pregnant women,22–25 homosexual men,26 commercial sex workers,10,27,28 and STD clinic29 and substance abuse30,31 patients. Since many of these studies relied on convenience sampling, estimated rates of HSV-1/HSV-2 infection and co-infection may have been subject to self-selection bias.
Our study findings suggest distinct etiologies for HSV sub-types, although HSV-1 and HSV-2 may co-exist and share socio-demographic and behavioral risk factors. The nearly two-fold increased odds of HSV-2 infection among females reinforces the idea that sexual activity plays a key role and that females are more vulnerable than males to acquiring this infection through the genital route. The lack of trend by age implies that environmental exposures during childhood may be more salient than later exposures for the acquisition of HSV-1 infection. Specifically, HSV-1 infection could be attributed to poor hygienic conditions during childhood, eventually leading to non-genital forms of herpes.1 Acquisition of HSV-1 during childhood may protect, in part, against later acquisition of HSV-2 or reduce the likelihood of clinical manifestations associated with HSV-2 infection.1 Nevertheless, the steady increase with age of HSV-1/HSV-2 co-infection is expected as it is consistent with cumulative exposure to either strain over time. A cohort effect may also be operating whereby older individuals were more likely to have been infected with HSV-1 during childhood than their younger counterparts. Disparities by race1,5,32 and level of education1 are reminiscent of socio-cultural factors, namely poverty and patterns of sexual activity, as important risk factors for acquiring either type of HSV infection. While socioeconomic conditions may help explain elevated HSV-1 infection rates among Hispanics, both socioeconomic and behavioral risk factors might be playing a role in the high HSV-1 and/or HSV-2 infection rates of non-Hispanic blacks. Tobacco smoking,1,33 recreational drug use,34,35 and multiple sexual partners1,36 are risk behaviors that have been previously linked with numerous STDs, including HSV-2. Our results suggest that HSV-2 infection and HSV-1/HSV-2 co-infection are consistently influenced by these behaviors, most of which are characteristic of low socioeconomic conditions. Interestingly, lifetime use of alcohol was not consistently related to HSV-1/HSV-2 infection and co-infection. This result is not aligned with studies of STDs and alcohol use disorders.31,38,39 It is plausible that, unlike the use of tobacco and recreational drugs, which are clearly addictive behaviors affecting mainly sensation seekers,40,41 identifying a role for alcohol use in STDs depends on how it is defined and measured. A recent systematic review implied that the overall evidence for an effect of alcohol consumption on STD acquisition remains equivocal due to methodological differences among existing studies.42 Instruments that measure event-specific alcohol use or alcohol abuse and dependence rather than usual patterns of consumption, may be better able to discriminate individuals at high risk for acquiring HSV and other STDs. Alternatively, the lack of association between alcohol consumption and HSV infection may be the result of confounding by other measured and unmeasured factors.
Our results should be interpreted with caution. First, given the cross-sectional design, a temporal relationship between risky behaviors and HSV-1/HSV-2 infection cannot be established. For the same reason, we cannot ascertain the sequence of events leading to genital herpes. Second, measurement error is an issue especially for self-reported demographic, socioeconomic, and risk behavioral characteristics. Since HSV-2 is clearly acquired through sexual activity, the fact that some of those who tested positive for HSV-2 reported themselves as abstainers raises concerns about the validity of questionnaire items and/or HSV-2 testing. It is worth noting that more advanced and commercially available methods for detecting HSV-1/HSV-2 infection such as the University of Washington Western blot assay and Focus Technologies tests have emerged since the NHANES data were collected. Third, we relied on existing data and were, therefore, unable to include refined exposure measurements.
In conclusion, HSV-1 infection appears to be predominantly influenced by low socioeconomic status and may be acquired at any time through various routes, genital or otherwise. Conversely, HSV-2 seropositivity was associated with sexual behavior, tobacco use and recreational drugs, in addition to demographic cofactors. The data show a more than two-fold increased association between sexual risk behavior and HSV-2 seropositivity, consistent with previous studies supporting the sexual transmission of HSV-2. Furthermore, genital herpes may be the outcome of either HSV-1 or HSV-2 infection. In the absence of effective vaccination, risk factors should be used to target specific sub-populations with health education and screening interventions, in an effort to prevent transmission to other individuals, including sex partners and offspring. Since both HSV sub-types are linked with genital herpes and associated health problems, continued efforts at vaccine development should be expanded to include both HSV-1 and HSV-2 sub-types.
Acknowledgments
No funding was provided for this project. However, this research was supported in part by the intramural research program of the NIH, National Institute on Aging. We would like to thank the Brickell Library at Eastern Virginia Medical School for providing access to peer-reviewed journals.
Footnotes
Presented in part at the 42nd Annual Society for Epidemiology Research Meeting, Anaheim, California, USA, June 23–26, 2009.
Conflict of interest: No conflict of interest to declare.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

