Sex education and STI fatalism, testing and infection among young African American men who have sex with women

Abstract

The purpose of this study was to examine the association between institution-delivered sex education given under real-world conditions and sexually transmitted infection (STI) rates, STI fatalism, and prior STI testing among African American men aged 15–24 who have sex with women. Participants were tested at community venues for Chlamydia and gonorrhoea and undertook a survey to elicit history of sex education and sexual health information. Among 1196 participants, 73.0% reported having received institution-delivered sex education topics including STI information (90.5%), condoms (89.2%), pregnancy/birth (72.1%) and birth control (67.1%). Among a subset of participants asked about the quality of sex education, 85.7% reported it was ‘very good’ or ‘OK’. Prevalence rate for Chlamydia and/or gonorrhoea was 10.5%. Those who received sex education were more likely to have lower STI fatalism (51.0% vs. 42.4%, p=0.01) and more likely to report previous Chlamydia screening (44.1% vs. 31.6%, p<0.01), but did not have a significantly lower rate of Chlamydia and/or gonorrhoea (9.9% vs. 12.4%, p=0.20) compared to those who did not receive sex education. These findings suggest that institution-delivered sex education given under real-world conditions has beneficial effects on STI risk factors among young African American men.

Introduction

African American youth have disproportionately high rates of sexually transmitted infections (STI) including Chlamydia trachomatis (Chlamydia) and Neisseria gonorrhoeae (gonorrhoea) (Centers for Disease Control and Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention 2019). In 2018, the rate of Chlamydia in the New Orleans-Metairie, Louisiana Metropolitan Statistical Area was 1.5 times the national average, at 827.6 cases per 100,000 people, and the rate of gonorrhoea was 1.6 times the national average, at 280.9 cases per 100,000 people (Centers for Disease Control and Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention 2019).

The rate of Chlamydia among Black and African American residents in Louisiana in 2017 was 5.0 times higher than the rate among White/Caucasian peers and 2.9 times higher than Hispanic/Latinx peers (State of Louisiana Department of Health Office of Public Health 2018). Additionally, African American youth report engaging in behaviours that increase risk of Chlamydia and gonorrhoea infection more frequently than White and Hispanic peers, including having ever had sex, having four or more partners during their lifetime, being currently sexually active, and being less likely to have used a condom the last time they had sex (Centers for Disease Control and Prevention 2018).

Reviews of randomised controlled trials and studies using quasi-experimental designs that have assessed comprehensive sex education programmes show strong evidence that comprehensive sex education can effectively delay initiation of sexual intercourse, reduce sexual risk behaviours in adolescents, and reduce acquisition of STIs including HIV, and teenage pregnancy (Chin et al. 2012; Johnson et al. 2011; Denford et al. 2017). Most teenagers receive their formal reproductive health information from school or other institutions, but much of it is not comprehensive in character (Guttmacher Institute 2016) and the percentage of sex education programmes for boys that include birth control and other important topics is reported to be declining (Lindberg, Maddow-Zimet, and Boonstra 2016).

Although knowledge about sexual health and STIs has been shown to reduce sexual risk behaviours (Lee, Cintron, and Kocher 2014), only 29 states in the USA mandate sex education in schools (Guttmacher Institute 2020). Louisiana allows sex education but does not mandate it. Louisiana state law RS 17:281 (section on instruction in sex education) does not require sex education to be taught at any grade level, but it does allow it in grades 7–12 (students ages 12 to 18) statewide and starting in 3^rd grade (students ages 8 to 9) in New Orleans (2011 Louisiana Laws). In Louisiana, the sex education must emphasise abstinence but can include discussion of risk reduction methods like contraception and condoms. This law also prohibits surveying students about sexual beliefs and behaviours in school settings, creating a lack of sexual health data on this population (2011 Louisiana Laws).

Certain behaviours, beliefs and attitudes are associated with increased risk of STI acquisition, and comprehensive sex education has been shown to improve many of these factors (Chin et al. 2012; Denford et al. 2017; Johnson et al. 2011). It is unknown if the sex education given under real-world conditions being received by young African American men in New Orleans is achieving this positive change in risk behaviours and attitudes including increasing STI testing and lowering STI fatalism. STI prevalence among teenagers and adolescents could be reduced through STI testing due to counselling and STI awareness provided at time of testing or because infections are being identified and treated earlier. More knowledge about sexual health could also lead to lower STI fatalism. STI fatalism is the belief that one is destined to contract an STI and therefore taking steps to prevent disease is useless (Meyer-Weitz 2005). High STI fatalism may increase STI acquisition as people may feel there is no use in trying to protect themselves. Sex education could affect STI rates by impacting these inter-related beliefs and behaviours.

STIs such as Chlamydia remain under-studied among young African American men in Louisiana, despite persistent racial disparities in disease rates. The purpose of this study was to describe institution-delivered sex education received by young African American men in New Orleans including participant characteristics and to examine the associations between prior receipt of sex education and infection with Chlamydia and/or gonorrhoea, STI fatalism and prior STI testing.

Materials and Methods

Check It is a seek, test, and treat, community-based Chlamydia and gonorrhoea screening programme for African American men aged 15–24 years who have sex with women. This study is based on the concept that screening and treating men for Chlamydia and gonorrhoea in high STI prevalence areas can reduce rates among women in these communities (Gopalappa et al. 2013; Kissinger et al. 2014). Participants were recruited through venue-based recruitment and social marketing between May 2017 and February 2019. Recruitment sites included barbershops, minority-serving universities/colleges, job training programmes, health fairs and other community venues that young African American men frequent. Both active and passive methods were used for recruitment including staffing events at selected venues, referrals from community partners, social network peer referral, flyers, social media, and a recruitment page on the programme website (www.gocheckit.net).

Participation was limited to those who (1) were assigned male at birth (identified by asking “do you have a penis?”), (2) identified as African American or Black, (3) were 15–24 years of age, (4) lived or spent most of their time in Orleans Parish (county), (5) had vaginal sex with a woman in the past 60 days, (6) were able and willing to consent, (7) could speak and understand English, (8) had not taken azithromycin in the past 2 weeks, and (9) had not previously enrolled in Check It.

After eligible participants provided informed consent, a survey was administered, a first catch urine sample was obtained, and contact information was documented. Participants received modest monetary compensation ($20 equivalent) in the form of a voucher for services at their recruitment site (e.g. haircut, food) or a gift card. STI status was assessed using nucleic acid amplification test (NAAT) Roche cobas®, which tests for both Chlamydia and gonorrhoea urine specimens.

Surveys were conducted with the audio computer-assisted self-interview (ACASI) software QDS™ Questionnaire Development System on laptops and tablets, with participants having the option of using audio. The survey included questions on demographic characteristics, socio-economic status, sex education experiences, neighbourhood environment, symptoms, STI history, STI fatalism and health and sexual behaviours including partner-specific questions. Study data were managed in REDCap hosted by Tulane University (Harris et al. 2009).

Participants were asked if they had ever received sex education in school, after school, in church, or at some other location (i.e. institution-delivered). If they answered yes, then they were asked to indicate which sex education topics they had received out of a list of twelve. This list included STDs/STIs, condoms, birth control, sexual consent, abstinence, male anatomy, female anatomy, gender identity, positive/negative relationships, pregnancy/birth, fathering and sexuality. This list was selected to offer a comprehensive range of topics, derived from the Future of Sex Education (FOSE) list of core topics (The Future of Sex Education Initiative 2012). The number of topics reported by each participant was summed and participants were grouped by those reporting no institution-delivered sex education and by those reporting any institution-delivered sex education (one or more topics). Participants enrolled between 3 August, 2018 and 7 February, 2019 also completed an additional section of questions about perceived quality of institution-delivered sex education.

The STI fatalism score was calculated from a 5-item adapted version of a previously described HIV fatalism scale (Meyer-Weitz 2005; Bland et al. 2012). The term HIV was replaced with STI within each item statement. The 5 items were (1) ‘if one was meant to be infected with a sexually transmitted infection (STI) then there’s nothing you can do about it’, (2) ‘most of my friends think that getting an STI sooner or later is unavoidable’, (3) ‘most of my friends believe they can do something to prevent STIs’, (4) ‘if I were to be infected by an STI it will eventually be cured without any treatment’, and (5) ‘I believe I am at high risk for contracting an STI’. Responses were ‘strongly agree’, ‘agree’, ‘disagree’, and ‘strongly disagree’ with response scores ranging from 0 to 3. Scores for items 1, 2, 4 and 5 were inverted so that a higher number coded for high fatalism for all five items. Missing data were imputed using item-specific means. Among the 1196 participants, 26 (2.2%) were missing a value for the first item, 29 (2.4%) for the second, 29 (2.4%) for the third, 27 (2.3%) for the fourth, and 27 (2.3%) for the fifth. A composite score was calculated for each participant by summing the responses to the five items; the possible range was 0–15, with 0 being low fatalism and 15 high fatalism. The median composite score was used as a cut-off to classify participants as lower or higher fatalism as previously described for the HIV fatalism scale from which our STI fatalism scale was adapted (Bland et al. 2012).

Other variables included in this analysis were assessed in the same survey and are shown in Table 1. Prior screening for Chlamydia, HIV and other STIs (including gonorrhoea, herpes, syphilis, and warts, and other) were elicited. Race was categorised as one race or two or more races due to so few participants reporting a second race (all identified as African American in line with eligibility criteria). Binge drinking was defined as five or more drinks in a two-hour period. Participants were asked to select all drugs used in the past 2 months from a list of 9 options. Drug use was categorised as no recreational drugs, marijuana only, other recreation drugs, and refused to answer. Participants were asked to indicate all government assistance programmes received by anyone in their household in the past twelve months from a list including food stamps, purchase card, two Louisiana supplemental nutrition programmes (the Supplemental Nutrition Assistance Program for Women, Infants, and Children), Section 8 housing (rental housing subsidy), or other.

Table 1.

Characteristics of enrolled men and odds ratios for receipt of institution-delivered sex education compared to no institution-delivered sex education (n=1196)

Characteristic	Any institution-delivered sex education n=873	No institution-delivered sex education n=323	p-value	OR	95% CI
Age in years, mean (SD)	20.0 (2.5)	20.1 (2.5)	0.31	1.031	0.98–1.08
Race, n (%)			0.57
1 race (African American)	837 (95.9%)	312 (96.6%)		1.0	-
2 or more races2	36 (4.1%)	11 (3.4%)		1.22	0.61–2.43
Ethnicity, n (%)			0.79
Non-Hispanic, African American	846 (96.9%)	312 (96.6%)		1.0	-
Hispanic/Latino	27 (3.1%)	11 (3.4%)		0.91	0.44–1.85
School Completed, n (%)			0.24
Completed HS/GED	404 (46.4%)	161 (50.2%)		1.0	-
Still in HS	143 (16.4%)	47 (14.6%)		1.21	0.83–1.77
Did not complete HS/GED	23 (2.6%)	15 (4.7%)		0.61	0.31–1.20
Some college/trade school or above	299 (34.3%)	97 (30.2%)		1.23	0.92–1.65
Other	2 (0.2%)	1 (0.3%)		0.80	0.07–8.85
Vaginal sex partners in past 2 months3, n (%)			0.21
1 partner	589 (67.5%)	224 (69.4%)		1.0	-
2 partners	182 (20.9%)	54 (16.7%)		1.28	0.91–1.80
≥3 partners	102 (11.7%)	45 (13.9%)		0.86	0.59–1.26
Participants with male anal or oral sex partners in lifetime, n (%)	59 (6.7%)	25 (7.7%)	0.56	0.86	0.53–1.40
Alcohol consumption in past month, n (%)			0.08
Yes	415 (47.5%)	133 (41.2%)		1.30	1.00–1.69
No	451 (51.7%)	189 (58.5%)		1.0	-
Unknown	7 (0.8%)	1 (0.3%)		2.9	0.36–24.01
Binge drinking in past 2 months vs. none, n (%)	248 (28.6%)	80 (24.8%)	0.19	1.21	0.91–1.63
Recreational drug(s) in past 2 months4, n (%)			0.57
No recreational drugs	352 (40.3%)	136 (42.1%)		1.0	--
Marijuana only	431 (49.4%)	159 (49.2%)		1.05	0.80–1.37
Other recreational drugs	68 (7.8%)	24 (7.4%)		1.09	0.66–1.82
Refused to answer	22 (2.5%)	4 (1.2%)		2.13	0.72–6.28
Method of paying for clinic of doctor, n (%)			0.83
Medicaid	530 (60.7%)	188 (58.2%)		1.0	-
Free clinic, sliding scale, or self-pay	179 (20.5%)	74 (22.9%)		0.86	0.62–1.18
Private insurance	123 (14.1%)	43 (13.3%)		1.01	0.69–1.49
Other	29 (3.3%)	12 (3.7%)		0.86	0.43–1.71
Don’t know or refused	12 (1.4%)	6 (1.9%)		0.71	0.26–1.92
Participant or household member using government assistance programme(s), n (%)	327 (37.5%)	128 (39.6%)	0.49	0.91	0.70–1.19

Abbreviations: SD, standard deviation; HS, high school; GED, general education development.

¹Odds ratio is for a one-year change in age.

²All participants self-identified as African American, the reference group is one race (African American).

³All participants had at least one vaginal sex partner during the past 2 months to be eligible to participate, the reference level is one vaginal sex partner.

⁴Recreational drugs included marijuana, crack cocaine, cocaine, heroin, ecstasy/molly, drank/lean, crystal meth, prescription drugs.

Statistical analysis included bivariate analysis of selected patient characteristics by ‘any institution-delivered sex education’ and ‘no institution-delivered sex education’ with number, percent, and chi-square tests for categorical variables and mean, standard deviation, and t-tests for continuous variables. Frequency statistics for self-reported sex education include number with percent or mean with standard deviation. For tests of association between history of institution-delivered sex education and the three outcomes (including individual STI fatalism scale items), chi-square analysis was conducted, with Fisher’s Exact Test being used where expected cell counts were ≤5. The three outcomes include STI status at enrollment, previous STI testing, and STI fatalism composite score. Logistic regression was used to produce odds ratios and 95% confidence intervals for history of institution-delivered sex education by demographic and behavioural characteristics and the outcomes. All analyses were conducted using SAS 9.4.

While STI fatalism and number of topics reported from institution-delivered sex education were both measured with numeric scales, the decision was made a priori to analyse both dichotomously. This was done in order to group all possible curricula together under ‘any institution-delivered sex education’. STI fatalism score was analysed using the sample median, as previously described for the HIV fatalism scale from which the STI fatalism scale was adapted (Bland et al. 2012). The Cronbach’s alpha for this scale was 0.61, suggesting moderate reliability. To more fully explore the relationship these two measures had with other variables, the analysis was repeated using quartiles of both number of topics of sex education reported and STI fatalism composite score. Additionally, the relationship between each individual item in the STI fatalism scale and institution-delivered sex education is reported.

Results

Among 1607 people recruited for Check It during this time, 1241 met the eligibility criteria for enrollment. Among the eligible men 34 declined to participate. Eleven enrolled participants lacked complete data, resulting in 1196 men being included in this analysis. A subset of 309 participants also answered questions on perceived quality of sex education which were added later in the study. (Figure 1).

Figure 1.

Flow chart of participant recruitment and enrollment

Among all participants, the mean age was 20.0 years (s.d. 2.5) and all participants identified as African American with 3.9% identifying as two or more races. Hispanic ethnicity was reported by 3.2% of participants. While 47.4% of participants had a high school diploma or general education development (GED), and 33.2% had some college education, trade school education, or above, 15.9% were still in high school and 3.2% were no longer in school and did not complete a high school diploma or GED. When paying for a visit to a clinic or doctor’s office, 60.0% of participants reported using Medicaid, 21.2% used free clinics, sliding scale clinics, or self-paid, while 13.9% used private insurance. Just over 38% of participants reported that they or someone they lived with received money or services from government assistance programmes.

Nearly half (45.8%) of the participants consumed alcohol in the past two months and 27.6% reported at least one episode of binge drinking (having 5 or more drinks within a two-hour period). Over half (57.0%) reported any recreational drug use in the past 2 months; 49.3% used marijuana only, and 7.7% used other drugs including drank/lean (cough medicine-based drink) (4.2%), ecstasy/molly (1.4%), cocaine (0.7%), and/or other drugs (3.3%). In the two months before enrollment, all participants reported at least one vaginal sex partner per eligibility requirements, and 32.0% had two or more female partners. While 7.0% of participants reported ever having a male sex partner in their lifetime, 4.3% reported a male partner in the previous two months.

Bivariate analysis of demographic and behavioural characteristics by receipt of institution-delivered sex education are shown in Table 1. None of the selected demographic and behavioural characteristics considered in Table 1 differ by receipt of institution-delivered sex education.

Institution-delivered sex education topics and location are summarised in Table 2. Institution-delivered sex education was reported by 873 (73.0%) participants starting at a mean age of 13.7 (s.d. 2.2) years. Topics reported include STIs (90.5%), condoms (89.2%), pregnancy/birth (72.1%), birth control (67.1%), sexuality (64.0%), sexual consent (58.7%), male anatomy (56.0%), female anatomy (54.2%), abstinence (45.1%), positive/negative relationships (44.9%), gender identity (43.6%), fathering (37.3%) and other topics (4.4%). Participants reported receiving a median of 7 topics out of a possible 12 listed in the survey (mean=6.8, s.d.=3.3) and no participant reported abstinence as the only topic received from institution-delivered sex education. The majority of participants (84.5%) reported that institution-delivered sex education was received at a public school, while 5.4% from a church/religious group, 5.3% from a private school (non-religious), 3.9% from a religious school, and 5.2% from other sources. Among the 309 participants completing questions on sex education quality, 217 (70.2%) reported any institution-delivered sex-education, which was similar to the percentage among the total sample (73.0%). Responses to perceived quality of the sex education was mixed. Nearly half (49.8%) thought the quality was ‘OK,’ 35.9% thought it was ‘very good,’ and 7.8% thought it was ‘not very good.’ The final 6.5% of respondents did not have an opinion or refused to answer.

Table 2.

Self-reported institution-delivered sex education (n=1196)

Variable
Ever had institution-delivered sex education, n (%)	873	(73.0%)
Among those with any institution-delivered sex education:
Age first received institution-delivered sex education, mean (SD)	13.7	(2.2)
Sex education topics, n (%)
STDs/STIs	790	(90.5%)
Condoms	779	(89.2%)
Pregnancy/birth	629	(72.1%)
Birth control	586	(67.1%)
Sexuality1	242/378	(64.0%)
Sexual Consent	512	(58.7%)
Male anatomy	489	(56.0%)
Female anatomy	473	(54.2%)
Abstinence	394	(45.1%)
Positive/negative relationships2	222/495	(44.9%)
Gender identity	381	(43.6%)
Fathering	326	(37.3%)
Other	38	(4.4%)
Number of topics received, mean (SD)	6.8	(3.3)
Where institution-delivered sex education was received3, n (%)
Public school	738	(84.5%)
Church or religious group	47	(5.4%)
Private school (non-religious)	46	(5.3%)
Religion-based school	34	(3.9%)
Other	44	(5.2%)
Among the subset asked about quality of institution-delivered sex education:
Received institution-delivered sex education, n (%)	217	(70.2%)
Perceived quality of institution-delivered sex education, n (%)
Very good	78	(35.9%)
OK	108	(49.8%)
Not very good	17	(7.8%)
No opinion	13	(6.0%)
Refuse to answer	1	(0.5%)

Abbreviations: SD, standard deviation; STD, sexually transmitted disease; STI, sexually transmitted infection.

¹The sample size for this topic is lower due to the option being removed from the survey.

²The sample size for this topic is lower due to the option being added to the survey after data collection had started.

³Participant could check as many responses as applicable.

Tests for associations between sex education and the outcomes STI diagnosis at enrollment, history of screening and STI fatalism are shown in Table 3. No statistically significant difference was found for Chlamydia and/or gonorrhoea between participants with a history of institution-delivered sex education compared to those who received no institution-delivered sex education (9.9% vs 12.4%, p=0.20).

Table 3.

STI diagnosis at enrollment, history of screening, and STI fatalism by sex education (n=1196)

Outcome	Overall (n=1196)	Any institution-delivered sex education (n=873)	No institution-delivered sex education (n=323)	P-value	Odds Ratio	95% CI
STI status at study enrollment	n (%)	n (%)	n (%)
Chlamydia Positive	122 (10.2%)	83 (9.5%)	39 (12.1%)	0.19	0.76	0.51–1.14
Gonorrhoea Positive	14 (1.2%)	10 (1.2%)	4 (1.2%)	0.991	0.92	0.29–2.96
Chlamydia and/or Gonorrhoea Positive	126 (10.5%)	86 (9.9%)	40 (12.4%)	0.20	0.77	0.52–1.15
Previous STI screening
Chlamydia	487 (40.7%)	385 (44.1%)	102 (31.6%)	<0.01	1.73	1.32–2.27
HIV	691 (57.8%)	517 (59.2%)	174 (53.9%)	0.10	1.26	0.97–1.63
STIs other than chlamydia or HIV	388 (32.4%)	307 (35.2%)	81 (25.1%)	<0.01	1.63	1.23–2.18
Low STI fatalism using 5-item composite score2	582 (48.7%)	445 (51.0%)	137 (42.4%)	<0.01	1.41	1.09–1.83

Abbreviations: STI, sexually transmitted infection; HIV, human immunodeficiency virus; CI, confidence interval.

¹Fisher’s Exact Test used due to small cell count.

²Low STI fatalism is defined as those with a cumulative score below the sample median.

Those receiving institution-delivered sex education were statistically more likely to report a history of prior Chlamydia testing (44.1% vs. 31.6%, p<0.0001) and prior STI screening other than Chlamydia and HIV (including gonorrhoea, herpes, syphilis, warts, and/or hepatitis) (35.2% vs. 25.1%, p<0.001) compared to those without institution-delivered sex education. Those with institution-delivered sex education were not more likely to report prior HIV screening (59.2% vs. 53.9%, p=0.10) compared to those without institution-delivered sex education. (Table 3).

The median composite score for STI fatalism after imputation was 5 (inter-quartile range 3–6). Participants receiving institution-delivered sex education were significantly more likely to endorse lower STI fatalism (i.e. lower than the median) compared to those without institution-delivered sex education (51.0% vs. 42.2%, p<0.01) (Table 3). The responses to all five individual STI fatalism items indicated low endorsement of STI fatalism (Table 4). Four of the items on the STI fatalism scale were significantly associated with receipt of sex education, with a greater proportion of young men receiving any sex education reporting lower endorsement of STI fatalism.

Table 4.

Individual STI fatalism items by sex education (n=1196)

Individual STI fatalism items1	Overall (n=1196)	Any institution-delivered sex education (n=873)	No institution-delivered sex education (n=323)	P-value
Lower fatalism response to: If one was meant to be infected with a sexually transmitted infection (STI) then there’s nothing you can do about it.	910 (76.1%)	680 (77.9%)	230 (71.2%)	0.01
Lower fatalism response to: Most of my friends think that getting an STI sooner or later is unavoidable.	907 (75.8%)	672 (77.0%)	235 (72.8)	0.13
Lower fatalism response to: Most of my friends believe they can do something to prevent STIs.	1021 (85.4%)	749 (85.8%)	272 (84.2%)	0.49
Lower fatalism response to: If I were to be infected by an STI it will eventually be cured without any treatment.	987 (82.5%)	734 (84.1%)	253 (78.3%)	0.02
Lower fatalism response to: I believe I am at high risk for contracting an STI.	970 (81.1%)	722 (82.7%)	248 (76.8%)	0.02

Abbreviations: STI, sexually transmitted infection.

¹Responses indicating higher STI fatalism are grouped and compared to responses indicating lower STI fatalism. Responses include ‘strongly agree’, ‘agree’, ‘disagree’, and ‘strongly disagree’.

The sensitivity analysis measuring institution-delivered sex education as quartiles of number of topics received did not substantively change any of the associations found when a dichotomous measure was used. Similarly, when using quartiles of STI fatalism composite score, similar associations were found for all relationships examined.

Finally, neither the association between STI fatalism nor previous STI screening was significantly associated with Chlamydia and/or gonorrhoea positivity. Among participants with no prior STD test, 10.6% had Chlamydia and/or gonorrhoea while 10.8% of those with a prior STD test were positive (p=0.88). While 11.0% of participants with lower fatalism had Chlamydia and/or gonorrhoea, 10.1% of those with higher fatalism were positive (p=0.61). However, those with lower STI fatalism were significantly more likely to report prior STI testing compared to those with higher STI fatalism (38.0% vs. 28.2%, p<0.001).

Discussion

Most (73.0%) of the young African American men enrolled in Check It reported receiving institution-delivered sex education. Most had received this in school, and none reported receiving abstinence as the only topic included. As in many other states in the USA, sex education in schools is not mandatory in Louisiana; there is no consistent, comprehensive sex education curriculum across the state, leaving each local public or parish school board to decide if and how to implement it. Consequently, institutionally delivered sex education can be lacking, fragmentory and not evidence-based.

In this study, young African American men who reported receiving institutionally delivered sex education were more likely to have been tested for Chlamydia and STIs other than Chlamydia and HIV in the past and more likely to endorse lower STI fatalism. Those with institutionally delivered sex education did not, however, have a significantly lower rate of Chlamydia and/or gonorrhoea.

Our data suggest that institutionally delivered sex education may be influential in reducing STI fatalism. Fatalistic attitudes have been studied regarding beliefs about HIV and AIDS as well as pregnancy. In the context of HIV and pregnancy, a fatalistic attitude is often associated with risky sexual behaviours, including less frequent use of condoms and contraception by adolescents and adults (Edelman et al. 2015; Hardeman, Pierro, and Mannetti 1997; Frost, Lindberg, and Finer 2012). A fatalistic attitude may leave one feeling it is useless to try and protect oneself, creating a reluctance to change risky sexual behaviours (Meyer-Weitz 2005). Knowledge about sexual health has been associated with lower fatalism (Benghiac 2012; Hess and McKinney 2007) but the relationship between sex education programmes and STI fatalism among adolescents and young adults is not well studied. Better understanding of the source of STI fatalistic views among young African American men and the impact on risky behaviours are needed to improve intervention strategies for reducing STIs, including increasing STI testing. Further research is needed to understand how institutionally delivered sex education interacts with the other possible sources of a fatalistic attitude.

The temporal relationship between receipt of institutionally delivered sex education and Chlamydia and/or gonorrhoea positivity, previous STI screening, and STI fatalism would be of interest. It is possible that previous STI screening and STI fatalism are on the causal pathway between receipt of sex education and STI status at enrollment. Without the temporal relationship between these variables being clear, it is not possible to explore these relationships further. The data presented in Table 1 did not identify any variables as potential confounders. Longitudinal studies are required to more deeply explore the relationships between these factors and STI prevalence.

Limitations

There are several limitations to the study. First, the study recruited men by means of a convenience sample of sexually active young African American men residing in New Orleans. The sample was drawn from a wide geographic range of New Orleans, so it may be generalisable to the target population of young African American men in New Orleans. It is possible that young African American men with higher STI fatalism were less likely to participate in the Check It Programme. However, heterogeneity in STI fatalism composite scores were observed, suggesting men with both low and high STI fatalism participated. Second, the cross-sectional design employed in this study makes it difficult to determine the temporal relationship between sex education and the three outcome measures, so causal relationships cannot be determined. Third, self-reported outcomes are susceptible to social desirability bias. To minimise this potential threat, audio computer-assisted self-interview software was used, which has been shown to elicit more valid responses than other modalities (Kissinger et al. 1999; Christopher K. Fairley et al. 2010). Fourth, the quality of sex education was determined via participants’ subjective assessments and the actual quality is unknown. Finally, the 5-item STI fatalism scale employed in the study had been validated for attitudes toward HIV but was adapted for all STIs in this study.

Conclusion

This is one of the first studies to examine the association between institution-delivered sex education in real-world situations (rather than under ideal study conditions) among young African American men who have sex with women. The trends seen among participants in the study suggest that institution-delivered sex education under real-world conditions can have a positive effect on the sexual health of young African American men.