Adolescents and young adults 15 to 24 years of age represent approximately 25% of the sexually active US population, but account for half of all new sexually transmitted infection (STI) cases and have the highest incidence of Chlamydia trachomatis (CT) and Neisseria gonorrheae (GC) among any sexually active age group.1 A large proportion of these infections are asymptomatic; thus, diagnosis and prompt treatment are vitally important to prevent transmission and long-term morbidity.1
Previous literature has suggested the emergency department (ED) be considered as a potential venue for STI screening.2–4 Adolescents comprise 15% of all ED visits and rely heavily on emergency services for their health care.5 In many cases, the pediatric ED (PED) may be the only opportunity for STI screening in adolescents. Previous studies have examined rates of GC and CT in the ED setting and have found high rates (10%–20%) and screening acceptability (60%–80%).4,6–15 Most of these studies have focused on older teens/adults, have not excluded symptomatic patients, and/or have screened based on patient-reported sexual history.
Studies have examined the barriers to routine screening among adolescents in traditional settings, with privacy being one of the most common patient concerns.16 Many studies have surveyed adolescents about demographics, sexual behaviors, and health care use, but to our knowledge, there is a lack of literature regarding reasons for declining STI testing in the PED setting.
The primary objectives of this study were to determine the proportions of GC and CT infections in asymptomatic adolescents in an academic urban PED and to determine patient-related predictors of screening refusals.
MATERIALS AND METHODS
This was a cross-sectional study wherein we enrolled a convenience sample of adolescent males and females, 14–21 years of age, who presented to the PED of a children's hospital with a non–STI-related chief complaint between September 2012 and July 2013. Recruitment took place across a wide range of hours based on principal investigator (PI) and/or research assistant availability. The hospital is an urban, tertiary care, academic center with annual ED visits of approximately 90,000 patients. Adolescents comprise nearly 20% of these visits, with 49% white (vs. nonwhite) and 60% with government-issued or no insurance (vs. private insurance).
Females were excluded from the study if they presented with a chief complaint of lower abdominal pain, and both sexes were excluded if they presented with genitourinary complaints or if the clinical care team ordered STI testing as part of their ED evaluation. In addition, patients were excluded if they were unable to participate because of severity of illness or injury, altered mental status, ingestion of substances of abuse or alcohol, receipt of medications for sedation, or developmental delay, or were non–English speaking.
Participants were identified as potentially eligible by the study PI or research assistant after being triaged. Each participant was approached in a private room and informed that the study was about a confidential topic. In an attempt to enroll patients confidentially, any others present in the room were asked to step out before details of the study were discussed. If others refused to leave the room, the study was still presented. Participants were not screened based on sexual activity, but instead were informed that GC and CT infections can be asymptomatic, and therefore, it is recommended that anyone sexually active receive yearly screening. The patients were then asked whether they wanted to submit a urine sample for GC and CT screening. Both those who agreed to provide urine and those who refused were subsequently asked to complete a questionnaire regarding potential barriers for STI screening. Participants were enrolled in 1 of 3 ways: (1) only providing a urine sample, (2) only completing the questionnaire, or (3) completing both. A confidential telephone number for follow-up was obtained from participants who agreed to urine screening. Adolescent verbal informed consent/assent was documented. Demographic variables of those declining participation were collected. This study was reviewed and approved by the institutional review board at our hospital with a waiver of parental consent.
Urine STI Screening Methods
Urine was obtained from participants in a sterile specimen cup and transferred by pipette into a preservative transport kit within 8 hours of collection. The samples were stored at 2°C to 8°C for up to 3 days until shipped to the Ohio Department of Health's laboratory. At this site, the urine was tested for GC and CT using BD ProbeTec Qx Amplified DNA Assays (BD Diagnostic Systems, Baltimore, MD). The PI received all results from the Ohio Department of Health via secure fax. Within 1 week of test collection, participants with positive results were notified by the PI using the confidential telephone number provided. One week later, the PI assessed whether adequate treatment was sought.17
The questionnaire was designed to identify patient-related factors that may influence a patient's decision to refuse screening for GC and CT. Questions using a 5-point ordinal response scale covered topics such as sexual history, concerns about privacy/confidentiality, and negative reactions to a positive test result (see Appendix A). Based on participants' responses to initial questions, some participants received open-ended questions to gain a better understanding of their reasoning behind own perceived risk for STI and any additional barriers to agreeing to screening. Validated questions regarding factors influencing refusal of STI screening were not available; thus, questions were developed by the study team. The questionnaire was pretested with 7 adolescents using cognitive interviewing methods.18 These participants read each question with the PI, and verbal probing was used to assess their understanding.
The questionnaire was self-administered using a laptop computer in the participant's private room at a time when clinical care was not being performed, as to not disrupt care or prolong their stay. The PI or research assistant was present to address questions and to ensure that others in the room would not attempt to read the questions or the participants' responses. The data were collected using Qualtrics (Provo, UT) online survey software. To ensure anonymity, the questionnaires were not linked to urine screening results.
Urine STI Screening Statistical Analysis
Frequency distributions were developed for all categorical variables. χ2 Test and Fisher exact test were used to compare demographic characteristics of those who enrolled with those who declined to participate in any part of the study. The demographic characteristics included age, race, sex, insurance status, and having a primary care physician (PCP). Bivariable and multivariable logistic regression was used to develop unadjusted and adjusted odds ratios. Analyses were conducted to compare (1) participants who agreed to provide urine with those who refused and (2) participants testing positive for STI with those testing negative.
Questionnaire Statistical Analysis
We tested for differences in responses to survey questions between patients who refused and those who agreed to STI screening using χ2 tests and the Mann-Whitney U test for categorical and ordinal scale data, respectively. This analysis was conducted for the total sample and by sex. In addition, bivariable and multivariable logistic regressions were used to develop unadjusted and adjusted odds ratios. The dependent variable was agreeing to urine testing. The independent variables were sexual activity and perceived risk of STI, controlling for sex.
Two authors (K.S., M.F.) independently coded the responses to open-ended questions by themes. During this process, the text of each response was carefully reviewed, with each response receiving one or more thematic codes that captured the idea being expressed. This process continued until thematic codes reached a level of saturation or a point at which all incoming data readily fit into existing codes and the emergence of new themes ceased.19 After doing this independently, the investigators reviewed and resolved any disagreements among themes assigned for a given response.
Of the 1054 adolescents approached for the study, 719 (68%) participated (Fig. 1). Most of those who declined were generally uninterested in the study (69%). Smaller percentages felt too ill to participate (23%) or did not wish to complete the study due to time restraints (5%). Among the participants, 441 (61%) provided urine samples; however, only 403 of these samples were successfully processed and resulted (see Table 3 footnote). Demographic variables among those enrolled in any portion of the study versus those who declined participation are shown in Table 1. Those enrolled were slightly more likely to have public insurance than those who declined participation (58% vs. 51%, P = 0.05). There were no other significant differences among other demographic variables.
Urine STI Screening Results
Table 2 displays demographic characteristics among participants who agreed to urine STI screening vs. those who refused STI screening. The bivariable analysis showed that those who agreed to screening were more likely to be nonwhite (61.4% vs. 38.6%, P = 0.001) and publically insured (63.3%) versus privately insured (29.3%) or no insurance (7.5%). A comparison of adjusted and unadjusted odds ratios indicates that race and insurance are potential confounders based on the decrease in the odds ratios when both are included in the model.
Among the urine STI screening participants, 40 (9.9%) were positive for at least 1 STI. Thirty participants were positive for CT only, 5 for GC only, and 5 for both infections. Table 3 provides the demographic characteristics of those positive for an STI as compared with those who were negative. Logistic regression demonstrated that when controlling for all other demographic characteristics, race was significant with the odds of testing positive among nonwhite participants 5.90 times the odds for white participants.
Table 4 summarizes questionnaire responses among the participants. Sexual activity was associated with whether or not patients agreed to screening (P = 0.01). Compared with those who refused STI screening, those who agreed were more likely to have had intercourse without a condom (35.4% vs. 23.4%) and less likely to endorse sexual activity (42.4% vs. 54.3%). An analysis by sex revealed that this held true for females but not for males.
Perceived risk of having an STI was also associated with whether patients agreed to STI screening. Patients who agreed to screening reported a slightly higher possibility of testing positive. This was the case overall and for both males and females. Table 5 summarizes the results of an analysis to further explore perceived risk and sexual activity, controlling for sex. It shows that these variables are likely confounders given the decrease in odds ratios when both are included in the model.
Participants from each of the agree and refusal groups who responded “not at all possible” that they would test positive, but also responded that they had engaged in oral sex or intercourse, received an open-ended follow-up question asking them to describe the reason for their response. Among those who received this question (n = 219), 85 provided a response. The main themes included the following: “practicing safe sex” (n = 20), “being in a monogamous relationship” or “perceiving that their partner(s) don't have an STI” (n = 18), “recently tested for STIs” (n = 14), and “not sexually active” (n = 20). Participants who refused screening were asked a final open-ended question regarding any additional reasons that affected their decision. Only 9 responded to this question, with the inability to urinate being the most commonly mentioned.
The proportion of patients with GC or CT in our study of asymptomatic adolescents was 9.9%. Other studies have shown similar high proportions of GC and CT infection in adolescents visiting EDs for care, but not all have not excluded those with genitourinary symptoms.7 Screening adolescents and identifying and treating infections in the absence of symptoms may play a role in preventing community spread of infection as well as potential long-term sequelae.
We did not screen participants based on whether or not they were sexually active as we aimed to reach participants that may not be honest about their sexual history. Previous literature has shown that adolescents often provide inaccurate self-reported sexual histories.20,21 In a large study by Diclemente et al.,21 10% of those who tested positive for an STI reported no sexual intercourse in the prior 12 months and 6% reported no sexual activity ever. In addition, physicians can be inconsistent in obtaining a sexual history as well as to whom they offer STI screening.22
In our study, enrollees were more likely to have public insurance than those who declined. Participants who specifically agreed to urine STI screening were also more likely to be publically insured and nonwhite, which supports findings of a prior study at our institution; however, these variables are potential confounders.23 Given that there are much higher rates of STIs in the black population, these youth may be more aware of the importance of screening and therefore more willing to be screened. Participants who were positive for an STI were also more likely to be nonwhite, which is also consistent with current Centers for Disease Control and Prevention reported rates where the CT rate among adolescent black girls is 5 times that of white females, and the CT rate among adolescent black boys is 9.5 times the rate of white boys.1 Thus, our data further supports the known racial disparity among STI-positive patients. In addition, nearly 90% of those agreeing to STI testing identified a PCP. Therefore, there is opportunity to integrate STI testing into the primary care setting.
The current literature lacks consistent evidence for the screening of adolescent boys, but the Centers for Disease Control and Prevention does recommend screening males in high-risk settings and men who have sex with men. In this study, similar proportions of males and females were positive for an STI indicating the importance of screening males in this high-risk PED setting. In addition, we found that males were not more likely to agree to screening even if sexually active, making them an even more important target for screening efforts.
One of the main barriers to STI screening was patient-perceived lack of risk. Consistent with our findings, previous studies suggest that patients decline testing due to perceived low risk for infection.8,24–26 Our study differed in that participants who agreed to urine STI screening were overall more likely to be at increased risk for an STI (i.e., reporting inconsistent condom use). Controlling for sex, however, these are potentially confounding variables. There were patients who refused urine screening despite being sexually active. In a study of young women, Baird and Merchant27 found that women who reported high-risk factors for STI were not more likely to accept an offer of STI testing. Thus, this is a population where further efforts should be focused.
Contrary to what one might expect, presence of others in the room, including parents, was not associated with refusal of STI screening. Previous studies have shown that requiring parental consent decreases adolescent participation in STI research as well as use of sexual health care services.23,28 Monroe et al.12 found results similar to ours; parental presence did not significantly deter participation in screening. Similar to our study, they did not screen based on sexual history, and it is possible that because we did not ask about sexual activity in our enrollment processes, participants felt more comfortable agreeing to enroll despite parental presence. Overall, we found that confidentiality (including concern about others discovering that they were tested or had a positive result or the likelihood of a negative reaction to a positive result) was not a significant patient-perceived barrier to screening, which is incongruent with previous literature.24,25 Perhaps among our population, the stigma associated with STIs is lessening or these adolescents are more forthcoming about their sexual health.
There are several limitations to our study. The 68% overall enrollment rate and only 61% of those enrolled agreeing to urine screening may limit the generalizability of our findings. This was a convenience sample of patients, and although we attempted to enroll every eligible adolescent presenting to the PED during enrollment times, the results may not represent the actual proportion of STIs in this PED. It is possible, however, that because we did not screen based on sexual activity, the proportion could be even higher. There may have been some adolescents that agreed to the urine STI screening, although they have never been sexually active. This was a concession we were willing to make in hopes of reaching at risk adolescents who would not have agreed to screening if they had to first admit to being sexually active. The screening offered was free and therefore removed the potential barriers related to cost and concern about potential for parents to learn about the testing via explanation of benefits provided by insurance companies. Also, we assumed that patients who were eligible for the study were asymptomatic, although we did not specifically ask about symptoms and instead relied on chief complaint and the fact that the primary care team did not send STI testing. Finally, the questionnaire was not validated and survey data were obtained through patient report which may limit the interpretation of the results. Although we took every possible precaution to ensure confidentiality, the presence of parents/others during the completion of the questionnaire may have affected responses. The survey data were not linked to the STI testing results, to offer participants anonymity, but limited our ability to analyze screening barriers in connection with STI screening results.
The patients in this study were not presenting to the PED with an STI-related complaint, and despite this, 10% were found to be positive for GC and/or CT. Although this study supports increased STI screening in the PED setting, most participants did identify a PCP. Many adolescents, however, rarely see their PCP and instead rely on the PED for their care. Therefore, a PED visit may be the only opportunity for screening. It is important for PCPs who care for adolescents to be vigilant about offering STI screening in the primary care setting during yearly physicals and/or during sick visits which may be the only time adolescents seek care. A universal STI screening program in the PED, and perhaps a collaborative effort with local health departments, may be a way to reach patients who would not otherwise be tested. Future research needs to be done to determine how best to make adolescents aware of their personal STI risk to encourage compliance with screening in PED settings.
1. Centers for Disease Control and Prevention. 2013 Sexually Transmitted Diseases Surveillance. 2013. Available at: http://www.cdc.gov/std/stats13/
. Accessed September 19, 2015.
2. Miller MK, Mollen CJ, O'Malley D, et al. Providing adolescent sexual health care in the pediatric emergency department: Views of health care providers. Pediatr Emerg Care 2014; 30: 84–90.
3. Mehta SD. Gonorrhea and chlamydia in emergency departments: Screening, diagnosis, and treatment. Curr Infect Dis Rep 2007; 9: 134–142.
4. Uppal A, Chou KJ. Screening adolescents for sexually transmitted infections in the pediatric emergency department. Pediatr Emerg Care 2015; 31: 20–24.
5. Ziv A, Boulet JR, Slap GB. Emergency department utilization by adolescents in the United States. Pediatrics 1998; 101: 987–994.
6. Jenkins WD, Nessa LL, Clark T. Cross-sectional study of pharyngeal and genital chlamydia and gonorrhoea infections in emergency department patients. Sex Transm Infect 2014; 90: 246–249.
7. Jenkins WD, Zahnd W, Kovach R, et al. Chlamydia and gonorrhea screening in United States emergency departments. J Emerg Med 2013; 44: 558–567.
8. Mehta SD, Hall J, Lyss SB, et al. Adult and pediatric emergency department sexually transmitted disease and HIV screening: Programmatic overview and outcomes. Acad Emerg Med 2007; 14: 250–258.
9. Miller MK, Dowd MD, Harrison CJ, et al. Prevalence of 3 sexually transmitted infections in a pediatric emergency department. Pediatr Emerg Care 2015; 31: 107–112.
10. Monroe KW, Jones M, Desmond R, et al. Health-seeking behaviors and sexually transmitted diseases among adolescents attending an urban pediatric emergency department. Compr Ther 2007; 33: 120–126.
11. Silva A, Glick NR, Lyss SB, et al. Implementing an HIV and sexually transmitted disease screening program in an emergency department. Ann Emerg Med 2007; 49: 564–572.
12. Monroe KW, Weiss HL, Jones M, et al. Acceptability of urine screening for Neisseria gonorrheae
and Chlamydia trachomatis
in adolescents at an urban emergency department. Sex Transm Dis 2003; 30: 850–853.
13. Miller CA, Tebb KP, Williams JK, et al. Chlamydial screening in urgent care visits: Adolescent-reported acceptability associated with adolescent perception of clinician communication. Arch Pediatr Adolesc Med 2007; 161: 777–782.
14. Ahmad FA, Jeffe DB, Plax K, et al. Computerized self-interviews improve chlamydia and gonorrhea testing among youth in the emergency department. Ann Emerg Med 2014; 64: 376–384.
15. Goyal M, Hayes K, Mollen C. Sexually transmitted infection prevalence in symptomatic adolescent emergency department patients. Pediatr Emerg Care 2012; 28(12): 1277–1280.
16. Pavlin NL, Gunn JM, Parker R, et al. Implementing chlamydia screening: What do women think? A systematic review of the literature. BMC Public Health 2006; 6: 221.
17. Schneider K, Byczkowski T, Reed J. Treatment compliance among asymptomatic adolescents with sexually transmitted infections. JAMA Pediatr 2015; 169: 1065–1066.
18. Sudman SBN, Schwartz N. Thinking About Answers: The Application of Cognitive Processes to Survey Methodology. 1st ed. San Francisco: Jossey Bass; 1996.
19. Strauss AL, Corbin JM. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 2nd ed. Thousand Oaks: Sage Publications; 1998.
20. Rosenbaum JE, Zenilman J, Melendez J, et al. Telling truth from Ys: An evaluation of whether the accuracy of self-reported semen exposure assessed by a semen Y-chromosome biomarker predicts pregnancy in a longitudinal cohort study of pregnancy. Sex Transm Infect 2014.
21. DiClemente RJ, Sales JM, Danner F, et al. Association between sexually transmitted diseases and young adults' self-reported abstinence. Pediatrics 2011; 127: 208–213.
22. Goyal MK, Hayes KL, Mollen CJ. Racial disparities in testing for sexually transmitted infections in the emergency department. Acad Emerg Med 2012; 19: 604–607.
23. Reed JL, Thistlethwaite JM, Huppert JS. STI research: Recruiting an unbiased sample. J Adolesc Health 2007; 41: 14–18.
24. Marrazzo JM, Scholes D. Acceptability of urine-based screening for Chlamydia trachomatis
in asymptomatic young men: A systematic review. Sex Transm Dis 2008; 35(11 suppl): S28–S33.
25. Balfe M, Brugha R, O'Donovan D, et al. Young women's decisions to accept chlamydia screening: Influences of stigma and doctor-patient interactions. BMC Public Health 2010; 10: 425.
26. Wolfers M, de Zwart O, Kok G. Adolescents in the Netherlands underestimate risk for sexually transmitted infections and deny the need for sexually transmitted infection testing. AIDS Patient Care STDS 2011; 25: 311–319.
27. Baird J, Merchant RC. A randomized controlled trial of the effects of a brief intervention to increase chlamydia and gonorrhea testing uptake among young adult female emergency department patients. Acad Emerg Med 2014; 21: 1512–1520.
28. Reed JL, Huppert JS. Predictors of adolescent participation in sexually transmitted infection research: Brief report. J Adolesc Health 2008; 43: 195–197.
APPENDIX A: Questionnaire—questions analyzed
- What if you did agree to get tested today?a As far as you know, how possible is it that you would test positive for either gonorrhea or chlamydia?
Could you briefly describe the reason(s) you think it's not at all possible that you might have gonorrhea or chlamydia?b
How upset would you, yourself, feel if you did happen to test positive?
- 1) Not at all possible
- 2) Only slightly possible
- 3) Somewhat possible
- 4) Very possible
- 5) Extremely possible
Many people your age engage in sexual behaviors including oral sex, vaginal intercourse, or anal intercourse. We do not judge these behaviors as right or wrong, but it's possible to get gonorrhea or chlamydia from any of these. Have you ever engaged in any of the behaviors described above?
Have you ever engaged in intercourse (vaginal or anal)?b
How often—would you say—did you use a condom?b
- 1) Not at all upset
- 2) Only slightly upset
- 3) Somewhat upset
- 4) Very upset
- 5) Extremely upset
Who was present in the room when you were offered testing?
- 1) Never
- 2) Rarely
- 3) Sometimes
- 4) Usually
- 5) Always
When we offered to provide a test for gonorrhea and chlamydia today, no one else was in the room, and we assured you no one would know you agreed to be tested. We understand some people might still be concerned that others might find out. How concerned were you about that?b
- 1) Patient was alone
- 2) Friend(s)
- 3) Sibling(s)
- 4) Parent(s)
- 5) Other adult(s)
We assured you that the results of the test would only be shared with you. We understand some people might still be concerned about others finding out about the results. How concerned were you about that?
- 1) Not at all concerned
- 2) Only slightly concerned
- 3) Somewhat concerned
- 4) Very concerned
- 5) Extremely concerned
Imagine testing positive for either gonorrhea or chlamydia.a How likely is it that someone you know would react in a bad way? (Think about parents, friends, and siblings)
- 1) Not at all concerned
- 2) Only slightly concerned
- 3) Somewhat concerned
- 4) Very concerned
- 5) Extremely concerned
We've asked you about a number of reasons why someone might not agree to be tested for gonorrhea or chlamydia during their visit to the emergency department. Did any other reasons affect your decision to not get tested today?
- 1) Not at all likely
- 2) Only slightly likely
- 3) Somewhat likely
- 4) Very likely
- 5) Extremely likely
aThe wording of the question was slightly different for those who agreed to urine STI screening.
bThis question was only asked those who responded with (1) to the preceding question.