Why Do Young Women Get Tested for Sexually Transmitted Infections? Evidence from The National Longitudinal Study of Adolescent Health

Abstract

Purpose:

Public health officials promote sexually transmitted infection (STI) testing as a method to reduce the morbidity of STIs. The purpose of this study was to examine what factors are associated with STI testing among young women across various types of STIs and to compare relative influences of factors across models.

Methods:

A secondary data analysis of data from Add Health Wave III was conducted (n=2629). Explanatory factors highlighted in qualitative literature were operationalized and built into a logistic regression model used to predict testing for chlamydia, gonorrhea, syphilis, human papillomavirus (HPV), genital herpes, and HIV.

Results:

STI symptoms and concerns about a recent sexual encounter were important expressed reasons for seeking medical care. Number of sexual partners, sexual orientation, STI symptoms, and going to the gynecologist in the past 12 months were important predictors of testing across STIs. This study supports qualitative work that suggests preventive health consciousness, STI symptoms, and relationship characteristics are important factors in STI testing. Results question the validity of self-report data regarding STI testing.

Conclusions:

Education efforts in secondary school health programs and during gynecologic examinations can decrease confusion about STI testing.

Introduction

Sexually transmitted infections (STIs) are one of the largest sources of morbidity for individuals aged 15–24 years old.¹ However, a majority of STIs go undiagnosed and untreated,² largely because of the asymptomatic nature of most STIs.^3

–6 As a result, public health emphasis has been placed on testing for STIs to identify asymptomatic and subclinical cases to prevent complications and to stop further proliferation.⁷

Testing can result in reduced prevalence and incidence of STIs in two ways. First, if an STI is curable, such as chlamydia infection, gonorrhea, or syphilis, testing provides a solution through treatment.³ Successful treatment of someone with one of these infections both cures the infection in the individual and stops future spread. If an STI is incurable, however, such as human immunodeficiency virus (HIV), herpes simplex virus (HSV), or human papillomavirus (HPV), testing may lead to a patient implementing behavioral changes to reduce further transmission of the virus to others.⁷ This may include abstaining from sexual intercourse or adopting safer sex practices, such as condom use, to reduce the risk of transmission to a partner.⁷ Treatment of curable infections and promoting behavioral change among individuals with incurable STIs can lead to a reduction of STI morbidity.

The Centers for Disease Control and Prevention (CDC) provides guidelines for routine STI testing practices. Currently, the CDC recommends routine annual testing for chlamydia for women <25 years old.⁷ Routine testing for gonorrhea is recommended for sexually active individuals who are at reasonable risk for infection.⁷ Women are also recommended to have routine Pap tests between the ages of 21 and 65 years.⁷ If atypical cells are noted in Pap tests results, it is recommended that cells then be tested for HPV.⁸ Routine screening for HIV among sexually active individuals is also recommended by the CDC, based on a growing body of literature that suggests HIV diagnosis is associated with significant and meaningful behavior change.⁹ On the other hand, routine serologic testing for syphilis and genital herpes is not currently recommended by the CDC for the general population.⁷ These guidelines may influence what STI tests a young woman is offered or recommended by a medical professional, influencing her choice in the decision to be tested for certain STIs.

Beyond current public health recommendations, individual patient characteristics play a role in STI testing. Qualitative evidence collected from women through interviews suggests that protecting personal health, as a check in between relationships, and experiencing symptoms of STIs are reasons for testing across types of STIs.¹⁰ However, quantitative studies exploring factors associated with reasons for STI testing across STIs are limited. Studies that examine individual characteristics for STI testing often do so on a disease-specific basis. Currently, no study examines determinants of STI testing across various types of STIs. Additionally, findings from existing qualitative literature have yet to be tested using a quantitative approach.

This study seeks to fill the gap in the literature regarding why young women are tested for STIs. The purpose of this study is to explore predictors of STI testing across a range of STIs and to make comparisons across models. By making comparisons across models, one can highlight factors that are STI-specific cues for testing or establish if the same factors are important across STIs. Focusing messaging on factors that are cross-cutting might be an efficient strategy for increasing wider STI testing rates.

Existing qualitative research has identified several themes about reasons for STI testing among young women. This study builds on these qualitative findings by using them as a theoretical basis for this quantitative study, using the key themes of preventive health consciousness, partner characteristics, and STI symptoms as the theoretical constructs that influence young women's decisions to be tested for STIs (Fig. 1). Using this model, one expects that those who have a greater preventive health consciousness, have a partner relationship characterized by higher risk perception for acquiring STIs (e.g., multiple partners), and have physical symptoms characteristic of an STI would be more likely to report testing. This is expected to vary across types of STIs. Public health recommendations regarding STI testing act as a filter through which STI testing occurs. Therefore, STIs that are recommended for routine testing for this age group (e.g., chlamydia) would have high rates of testing, whereas STIs that are not recommended for routine testing in this population (e.g., herpes) would have low rates of testing. It is expected that symptoms would play a larger role in testing among STIs not recommended for routine testing.

FIG. 1.

Diagram of the theoretical framework. STI, sexually transmitted infection.

Materials and Methods

Procedures

This study examines STI testing through secondary data analysis. Public use data from Wave III of the National Longitudinal Study of Adolescent Health (Add Health) were examined.¹¹ Wave I of Add Health used a complex, stratified cluster sample of U.S. schools to create a nationally representative sample of adolescents in grades 7–12. From the randomly selected school samples, students were randomly chosen to participate in an additional in-home questionnaire. During Wave III, students from Add Health Wave I in-home interviews were recontacted. Data were collected via in-home computer-assisted interviews between August 2001 and April 2002. The public use dataset contains 4882 randomly selected respondents from the contractual release dataset.¹² Of those, 2629 participants were women.

Participants

Participants' ages ranged from 18 to 28 years (mean 21.72, standard deviation [SD] 1.80). Non-Hispanic whites comprised the largest racial/ethnic group (58.5%), with 23.6% non-Hispanic black, 10.1% Hispanic, and 7.6% identifying as other race/ethnicity. Sexual orientation was skewed heavily toward 100% heterosexual (85.36%), although 9.7% identified as mostly heterosexual, 2.3% identified as bisexual, and 0.2% identified as 100% homosexual (gay). The majority of women had vaginal sex at least once in their lifetime (85.6%) and had some form of health insurance (79.3%).

Measures

The primary outcomes of interest were if participants had been tested for chlamydia, gonorrhea, syphilis, genital herpes, HPV, or HIV. Participants identified which, if any, sexually transmitted diseases (STD) they had been tested for in the past 12 months. Participants were asked to enter 1 in the space provided for any tests received in the past 12 months; each test had its own line. Individuals who did not place a 1 in the blank provided were coded as 0, not having received that test in the past 12 months. Answers of don't know, refused, and not applicable were recoded as missing. Each STI test had <1% missing data. Because of the binary nature of the outcomes of interest, logistic regression was used to model explanatory factors.

Several questions within the survey addressed reasons for seeing a doctor about STIs. Of those who saw a doctor or nurse because of STI concerns, questions about their reasons for doing so also where asked. The question asked: This most recent time, why did you think you might have a sexually transmitted disease (STD) or HIV? Mark all that apply. A similar 1–0 answering technique was applied to determine expressed reasons for seeking care.

The explanatory factors of preventive health consciousness, experiencing STI symptoms, and relationship characteristics from qualitative literature were operationalized using questions from the Add Health Wave III questionnaire. To assess one's preventive health consciousness, two variables were included in the model: having received a gynecologic examination in the past 12 months and having seen a doctor or nurse in the past 12 months. Both variables were recoded into binary variables, with 0=No and 1=Yes. Six questions were asked to capture the participant's experiences with STI-related symptoms in the past 12 months. Symptoms included painful or frequent urination, painful sores or blisters on the genitals, warts on the genitals, dripping or oozing from the vagina, bleeding after intercourse or between periods, and itching in the vagina or genital area. Similar to questions about STI testing, participants put a 1 in the blank if they had experienced this type of symptom in the past 12 months. Those who did not mark this box were coded as 0. Answers of don't know, refused, and not applicable were recoded as missing. Finally, relationship characteristics were measured through a combination of two variables. The first was the number of partners the participant engaged in vaginal intercourse with in the past 12 months, which was a count. Sexual orientation was also included and was measured using a 6-point Likert-type scale that ranged from 100% heterosexual (straight) to 100% homosexual (gay). Confounders examined included race/ethnicity, insurance status, and age. Race/ethnicity was grouped as Hispanic, non-Hispanic white, non-Hispanic black, and other, which were dummy coded for the logistic regression model. Insurance status was recoded into 1=Yes and 0=No. Age was a continuous variable calculated by subtracting the reported date of birth at Wave I from the current date.

Data analytic strategy

Data were analyzed using SAS version 9.2. Separate logistic regression models were conducted for each STI test of interest. All explanatory variables and confounders were included in each model. Logistic regression allows one to examine multiple predictors of dichotomous outcomes. By using the same variables in each model, one can compare and contrast the relative influence of each predictor variable across STI models. All variables were examined for significance at the p<0.05 level. This analysis was granted exempt status from the University of South Florida Institutional Review Board.

Results

Among this sample, 89.2% reported having seen a doctor or nurse in the past 12 months, and 74.2% indicated that they had gone to the gynecologist in the last 12 months. Testing rates varied across STIs. Chlamydia testing was the most commonly reported STI test received (23.7%) in the past 12 months, followed by gonorrhea (21.3%), HIV (20.9%), herpes (16%), syphilis (15.6%), and HPV (13.7%). Of the six tests analyzed in this study, women reported receiving from zero to six different STI tests in the past 12 months (mean 1.11, SD 2.00). Among women reporting being tested for chlamydia, 39.1% reported being tested for all six STIs. Similarly, 43.7% of those tested for gonorrhea, 59.5% of those tested for syphilis, 58.0% of those tested for herpes, 68.0% of those tested for HPV, and 44.4% of those tested for HIV reported being tested for all six STIs in the past 12 months.

Among women who had seen a doctor about STI concerns, various reasons prompted seeking healthcare. Reporting experiencing STI symptoms and worries about a recent sexual encounter were most predictive of being tested for various STIs (Table 1). Beyond expressed reasons for seeking medical attention, various predictors of STI testing derived from the qualitative literature are summarized in Table 2. Number of sexual partners in the past 12 months and having seen a gynecologist in the past 12 months were significant predictors of STI testing across all models. As the number of sexual partners increased, the likelihood of reporting testing for each STI increased. Among those who reported seeing a gynecologist n the past 12 months, the odds ranged from 1.89 (HIV) to 3.65 (gonorrhea and herpes) higher than those who did not report seeing a gynecologist in the past 12 months.

Table 1.

Women's Expressed Reasons for Seeking Medical Care for Sexually Transmitted Infections to Predict Type of Sexually Transmitted Infection Test Received

	Type of STI test ^a
	Chlamydia	Gonorrhea	Syphilis	HPV	Herpes	HIV
Physical symptoms	2.29 (1.30–4.04)^**	2.07 (1.20–3.58)^**	2.05 (1.18–3.56)^*	1.71 (0.97–3.01)	2.02 (1.16–3.53)^*	1.13 (0.66–1.93)
Partner has STI	3.07 (1.40–6.72)^**	1.98 (0.92–4.23)	2.44 (1.12–5.34)^*	1.97 (0.90–4.35)	1.22 (0.54–2.79)	1.79 (0.84–3.81)
Worried about recent encounter	2.52 (1.44–4.42)^**	2.17 (1.26–3.74)^**	2.64 (1.52–5.59)^***	1.50 (0.86–2.62)	2.08 (1.20–3.61)^**	1.93 (1.12–3.30)^*
Worried about other activities	1.11 (0.25–4.87)	1.32 (0.32–5.51)	1.25 (0.27–5.87)	1.50 (0.35–6.40)	3.93 (0.98–15.81)	1.31 (0.33–5.25)
Other	1.86 (0.94–3.67)	1.64 (0.84–3.21)	1.86 (0.93–3.72)	1.46 (0.72–2.94)	1.09 (0.53–2.21)	1.68 (0.87–3.24)

Questions asked only of those who were tested for sexually transmitted infections (STIs) in the past 12 months (n=516).

Adjusted odds ratios and 95% confidence intervals. Statistical significance determined by Wald's chi-square test.

p<0.05; ^** p<0.01; ^*** p<0.001.

HIV, human immunodeficiency virus; HPV, human papillomavirus.

Table 2.

Logistic Regression Results Displaying Differential Impact of Explanatory Variables Across Types of Sexually Transmitted Infection Test

	Type of STI test ^a
	Chlamydia	Gonorrhea	Syphilis	HPV	Herpes	HIV
No. of sexual partners in last 12 months	1.11 (1.04–1.18)^**	1.13 (1.06–1.21)^***	1.09 (1.02–1.15)^**	1.09 (1.03–1.15)^**	1.08 (1.02–1.14)^*	1.07 (1.01–1.13)^*
Sexual orientation	1.37 (1.07–1.59)^**	1.36 (1.11–1.66)^**	1.26 (1.01–1.57)	1.41 (1.12–1.76)^**	1.31 (1.05–1.62)^**	1.34 (1.10–1.63)^*
Gynecologist visit last 12 months	3.56 (2.51–5.04)^***	3.65 (2.52–5.30)^***	2.95 (1.97–4.40)^***	3.65 (2.28–5.83)^***	3.13 (2.09–4.70)^***	1.89 (1.38–2.59)^***
Saw doctor or nurse last 12 months	1.34 (0.84–2.14)	1.50 (0.90–2.51)	1.28 (0.74–2.18)	1.88 (0.95–3.73)	1.60 (0.90–2.85)	1.71 (1.06–2.76)^*
STI symptoms
Painful or frequent urination	1.79 (1.34–2.41)^***	1.53 (1.13–2.08)^**	1.55 (1.12–2.15)^**	1.85 (1.32–2.59)	1.38 (0.99–1.92)	1.78 (1.33–2.40)^***
Sores or blisters	1.28 (0.66–2.47)	1.53 (0.80–2.95)	2.51 (1.31–4.82)^**	1.64 (0.79–3.40)	4.63 (2.44–8.79)^***	1.48 (0.78–2.78)
Warts	2.12 (1.01–4.47)^*	1.82 (0.86–3.84)	1.44 (0.65–3.20)	5.89 (2.76–12.56)^***	2.18 (1.02–4.70)^*	1.18 (0.55–2.50)
Dripping or oozing	1.36 (0.80–2.32)	1.38 (0.81–2.36)	1.08 (0.60–1.95)	0.81 (0.41–1.60)	0.78 (0.41–1.46)	1.11 (0.65–1.89)
Bleeding between periods	1.30 (0.91–1.87)	1.36 (0.94–1.96)	1.39 (0.94–2.07)	1.16 (0.75–1.78)	1.16 (0.41–1.75)	0.99 (0.68–1.45)
Itching	1.15 (0.87–1.54)	1.19 (0.89–1.60)	1.00 (0.72–1.38)	0.98 (0.69–1.39)	1.26 (0.92–1.73)	1.27 (0.95–1.69)
Race/ethnicity
White	Referent	Referent	Referent	Referent	Referent	Referent
Black	1.37 (1.09–1.73)^***	1.34 (1.06–1.71)^*	1.34 (1.03–1.75)^*	1.05 (0.78–1.40)	1.20 (0.92–1.56)	1.44 (1.13–1.83)^**
Hispanic	0.97 (0.68–1.38)	0.95 (0.66–1.37)	0.93 (0.62–1.39)	0.93 (0.61–1.42)	0.76 (0.50–1.15)	1.19 (0.84–1.69)
Other	1.02 (0.69–1.51)	1.06 (0.71–1.59)	0.98 (0.62–1.54)	1.07 (0.67–1.71)	0.96 (0.62–1.51)	1.16 (0.78–1.73)
Age	0.95 (0.90–1.01)	0.97 (0.92–1.03)	0.93 (0.87–0.99)	0.99 (0.93–1.07)	0.94 (0.88–1.00)	0.97 (0.91–1.03)
Insured	1.06 (0.82–1.36)	1.05 (0.81–1.37)	0.90 (0.68–1.20)	0.92 (0.68–1.25)	0.94 (0.71–1.25)	1.02 (0.79–1.32)

Adjusted odds ratio point estimates and 95% confidence intervals. Significance assessed using Wald chi-square statistic.

p<0.05; ^** p<0.01; ^*** p<0.001.

The role of symptoms varied by type of STI. Those experiencing warts on the genitals were 5.89 (95% confidence interval [CI] 2.76-12.56) times more likely to report having been tested for HPV than those who did not report this symptom. Women experiencing sores or blisters were 2.51 (95% CI 1.31-4.82) times more likely to report being tested for syphilis and 4.63 (95% CI 2.44-8.79) times more likely to report being tested for herpes than those who did not experience sores or blisters. Symptoms typically associated with chlamydia and gonorrhea were not significant predictors, such as dripping and oozing. However, painful urination was a significant predictor of chlamydia, gonorrhea, syphilis, and HIV testing.

Discussion

There were several interesting patterns regarding differential predictors of STI tests. Expressed reasons for seeking medical care as a predictor of STI testing leaned heavily toward women reporting being worried about a recent encounter and experiencing symptoms. It is important to note that four of the five diseases reported as statistically significant associations with “worried about a recent encounter” are preventable through correct and consistent condom use.¹³

Relationship characteristics, such as number of sexual partners in the past 12 months and self-reported sexual orientation, were important explanatory variables across STI test types. Number of sexual partners was significant for every type of STI. The odds of getting tested ranged from 7% to 13% for each additional partner in the past 12 months. This corroborates with HIV testing literature. Samet et al.¹⁴ found that number of recent sexual partners was associated with an increased likelihood of STI testing. Genital herpes literature also supports this association.¹⁵ This may reflect an increased perceived risk of STI transmission and, therefore, increased STI testing. Even though sexual orientation showed significant associations with five of the six STI tests, the meaning of this is hard to determine. Results indicate that as one moves from reporting 100% heterosexual toward 100% homosexual, testing is more likely.

Those seeking preventive sexual and reproductive healthcare through an annual gynecologic examination were more likely than those who did not see a gynecologist in the past 12 months to be tested for STIs across the board. This is supported by findings from the HIV literature that suggests recently accessing sexual or reproductive services increases the likelihood of being tested for HIV.¹⁶ It seems that simply accessing medical services in general is not a trigger for STI testing, as indicated by the lack of association between having seen a doctor or nurse in the past 12 months and STI testing. The odds of STI testing as a result of seeing a doctor or nurse in the past 12 months were increased only for HIV testing. This could be because of the nature of HIV infection, which can often present as flu-like symptoms in its early stages, including fever, sore throat, swollen lymph nodes, and rashes.¹⁷ These types of symptoms may increase the likelihood of seeking care through normal health channels, as they do not clearly denote an STI. However, it cannot be concluded precisely why the patients accessed medical services in the past year.

The role of symptoms was also important in STI testing, especially for those with sores or blisters and warts. This supports existing research suggesting that experiencing symptoms is an important impetus for genital herpes testing.^18,19 It is interesting to note, however, that symptoms characteristic of chlamydia and gonorrhea, such as dripping or oozing and itching, did not play an important role in testing for these STIs. This could be because our sample was female. Women often do not experience symptoms related to these STIs, whereas men are more likely to report symptoms.²⁰

Race appears to be an important confounder with regard to STI testing. Black women were more likely than white women to report receiving testing for chlamydia, gonorrhea, syphilis, and HIV. Notably, however, black women were not significantly more likely to report testing for herpes and HPV despite having an increased burden of these STIs. As modeled, the lack of difference could be explained by the important role of symptoms in reporting testing, as variance associated with reporting visual signs of warts and blisters may obscure the variance associated with racial differences.

Findings from this study call into question the validity of self-report data for STI testing. Among women tested for STIs, a large majority indicated being tested for all six STIs within the past 12 months. This is highly unlikely when considering STI testing recommendations put forth by the CDC, especially regarding genital herpes. Testing for genital herpes is recommended only when symptoms are present.⁷ Thus, it is unlikely that 58% of women tested for any STI were also tested for herpes. A general trend toward checking off each STI is noted in the data if the individual was tested at all. Head et al.²¹ also found flaws in self-report data for STI testing. Among college women questioned after an annual well-woman examination, 25.7% falsely perceived they were tested for STIs during their gynecologic visit, and only approximately one in seven could accurately identify which STI they were tested for. This potentially highlights a low level of knowledge among young women about what STIs they were tested for when they “got tested.” This should be explored further in future studies through comparing self-report to medical records.

HPV was the least reported STI test received. Overall, 13.7% of the sample reported being tested for HPV. However, 82.4% of the overall sample indicated having a Pap test in the past 12 months, and 92% of the women who reported going to the gynecologist in the past 12 months indicated having a Pap test. Pap tests are a routine cervical cancer screening tool that detects atypical cervical cells. HPV is the causative agent for the majority of cancerous or precancerous cervical cell abnormalities. Therefore, it is recommended that atypical Pap tests are further analyzed for HPV DNA.⁸ Even though Pap tests do not directly measure HPV and are not specifically an HPV STI test, they can serve as a vehicle for identifying high-risk HPV disease. Thus, even though women are not being directly tested for HPV, they are being identified via Pap testing through testing of cell abnormalities.

Themes identified in qualitative research about reasons for STI testing seem to be important predictors of STI testing. Although women with increased numbers of sexual partners, increased healthcare use, and STI symptoms were more likely to report testing, some inconsistencies with theoretical underpinnings were also observed. Despite recommendations for routine Pap testing and against genital herpes testing, these rates were not reflected in the data. This could indicate a limitation of this model or could be due to validity of self-report data. Similarly, those identifying as <100% heterosexual were more likely to report STI testing in many categories. Consistent with this theoretical framework, sores or blisters were the most important predicator of herpes testing. The utility of this model should be explored further using primary data collected to increase conceptual accuracy.

Limitations

There were several limitations to this study, some of which are inherent to secondary data analysis itself. First, the analysis was limited to questions already existing in the Add Health Wave III questionnaire to operationalize themes proposed in the qualitative literature. Therefore, some concepts should be further developed in future studies to advance the utility of this framework with regard to a theoretical approach to STI testing. Next, the data were both cross-sectional and self-report. This limits the ability to make generalizations across time and may reduce the reliability or validity of the data collected. Because of the structure of the questionnaire, potentially important covariates, such as marital status, were not included in the model. Additionally, missing data caused the removal of approximately 18% of the respondents in the logistic regression analyses. The logistic regression models for gonorrhea and syphilis suggested poor data-model fit according to the Hosmer and Lemshow goodness-of-fit test. On the other hand, one of the primary goals of the article was to assess differences between tests, not to predict outcomes. Therefore, adapting each model would not be appropriate to answer this research question. Finally, it is important to put the current findings within an appropriate historical context. Data were collected before licensure of the HPV vaccine and the availability of many HPV DNA tests. Despite these limitations, this is the first quantitative study to examine the differential reasons for STI testing across all six of these STIs.

Conclusions

This study illuminates the need for health education in two distinct ways. These data clearly indicate that having a gynecologic appointment in the past 12 months was an important predictor of STI testing across the board. It is important for health educators in secondary school programs to highlight the importance of annual gynecologic examinations for maintaining both sexual and overall health. Additionally, secondary health education programs and units can focus on recognizing STI symptoms. Experiencing STI symptoms was an important predictor of being tested for certain STIs. It is important to recognize the symptom as STI related in order to get tested and, if necessary, treated.

Health education efforts are also necessary within the context of the gynecologic visit. Many women indicated that they were tested for everything. However, current STI testing guidelines promoted by the CDC do not recommend routine testing for all STIs. This disconnect between commonly accepted STI testing norms and perceptions of being tested for all STIs may indicate that young women are not clear about what is tested for during their well-women visits. Healthcare professionals need to clearly communicate what types of tests are available, and tests pertinent to the patient's age range and STI risk profile according to the CDC guidelines⁷ should be recommended. Clarification is necessary about what tests are actually being routinely performed. This offers a great opportunity for healthcare professionals to educate a captive audience.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Centers for Disease Control and Prevention. Sexually transmitted diseases in the United States, 2008: National surveillance data for chlamydia, gonorrhea, and syphillis, 2009. www.cdc.gov/STD/stats08/2008survFactSheet.PDF. 2010 February 17.

American Social Health Association. Sexually transmitted diseases in America: How many cases and at what cost? 27. www.ashastd.org/pdfs/std_rep.pdf

Centers for Disease Control and Prevention. CDC fact sheet: Chlamydia, 2007. www.cdc.gov/std/Chlamydia/Chlamydia-Fact-Sheet.pdf. 2010 February 17.

Farley

, Cohen

, Elkins

. Asymptomatic sexually transmitted diseases: The case for screening. Prev Med, 2003; 36:502–509.

Kamb

, Newman

, Peterman

et al. Most bacterial STD have no symptoms. International Conference on AIDS, 2000.

Cates

. Estimates of the incidence and prevalence of sexually transmitted diseases in the United States. Sex Transm Dis, 1999; 26:S2–S7.

Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2006. MMMR Recomm Reg, 2006; 55:1–94.

Wright

, Cox

, Massad

, Twiggs

, Wilkinson

. Consensus guidelines for the management of women with cervical cytological abnormalities. JAMA, 2002; 287:2120–2129.

Branson

, Handsfield

, Lampe

et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR, 2006; 55(RR14).

10.

Canso

. Young women and sexually transmitted infection testing: Investigating the impact of clinical experiences and social context on sexual health decision making. Masters thesis, Department of Health Care and Epidemiology, University of British Columbia, 2004.

11.

Udry

. The National Longitudinal Study of Adolescent Health (Add Health), Wave III, 2001–2002. Los Altos, CA: The Carolina Population Center, 2003.

12.

McKean

, Card

. The National Longitudinal Study of Adolescent Health (Add Health), Wave III, 2001–2002: A user's guide to the machine-readible files and documentation (Data Set Q1-Q9) Los Altos, CA: Sociometrics Corporation, 2003.

13.

Centers for Disease Control and Prevention. Male latex condoms and sexually transmitted diseases, 2009. www.cdc.gov/condomeffectiveness/latex.htm. 2009 February 20.

14.

Samet

, Winter

, Grant

, Hingson

. Factors associated with HIV testing among sexually active adolescents: A Massachusetts survey. Pediatrics, 1997; 100:371–377.

15.

Zimet

, Rosenthal

, Fortenberry

et al. Factors predicting the acceptance of herpes simplex virus type 2 antibody testing among adolescents and young adults. Sex Transm Dis, 2004; 31:665–669.

16.

Swenson

, Rizzo

, Brown

et al. Prevalence and correlates of HIV testing among sexually active African American adolescents in 4 US cities. Sex Transm Dis, 2009; 36:584–591.

17.

The Mayo Clinic. HIV/AIDS: Symptoms. www.mayoclinic.com/health/hiv-aids/DS00005/DSECTION=symptoms. 2010 May 5.

18.

Song

, Dwyer

, Mindel

. HSV type-specific serology in sexual health clinics: Use, benefits, and who gets tested. Sex Transm Infect, 2004; 80:113–117.

19.

Richards

, Krantz

, Selke

, Wald

. Healthcare seeking and sexual behavior among patients with symptomatic newly acquired genital herpes. Sex Transm Dis, 2008; 35:1015–1021.

20.

Centers for Disease Control and Prevention. Chlamydia: CDC fact sheet, 2007. www.cdc.gov/std/Chlamydia/STDFact-Chlamydia.htm. 2009 March 23.

21.

Head

, Crosby

, Shrier

, Moore

. Young women's misperceptions about sexually transmissible infection testing: A “clean and clear” misunderstanding. Sex Health, 2007; 4:273–275.