Added Value of Extragenital STI testing in “I Want The Kit” Program Users

Abstract

Background

The 2021 CDC STI treatment guidelines recommend extragenital testing for gonorrhea and chlamydia in men who have sex with men (MSM) and for women based on reported behaviors and exposures. The “IWantTheKit (IWTK)” program is a free online platform for specimen self-collection and mail-in for combined chlamydia/gonorrhea testing. We sought to assess the additional diagnostic value of extragenital testing compared to genital testing only for chlamydia/gonorrhea and determine factors associated with a positive extragenital test among IWTK users.

Methods

From August 2013 to January 2022, 7,612 unique IWTK users returned swabs for testing; 3,407 (45%) users requested both genital and extragenital tests and were included in this analysis. Descriptive statistics summarized demographic characteristics, reported behaviors, and genital and extragenital test results, data were stratified by gender and age group. A logistic regression model was used to estimate associations between factors and extragenital STI positivity.

Results

Chlamydia positivity was 4.7%, 2.4% and 1.5% at genital, extragenital and both sites were positive, respectively; for gonorrhea, 0.4%, 1.1% and 0.4% were positive at those sites, respectively. Among women, age 25 and younger was significantly associated with extragenital chlamydia, OR = 4.0 (p=0.010). Being in high risk quiz score group was associated with extragenital chlamydia (OR=2.6, p = 0.005), and extragenital gonorrhea in men and women (OR=8.5, p = 0.005).

Conclusion

Extragenital testing detected additional chlamydia and gonorrhea cases in the IWTK user population that would have been missed by genital only testing, especially for females <25 years and people reported high risk.

Short summary

A study in IWTK online STI testing platform users found that young women and people with high risk score were at higher risk of extragenital STIs.

Introduction

Sexually transmitted infections (STIs) rates are increasing in the US population; in 2018 the Centers for Disease Control and Prevention (CDC) estimated that 1 in 5 people had an STI.[1] Incidence of chlamydia increased from 478.8 to 552.8 cases per 100, 000, and the incidence of gonorrhea increased from 123.9 to 179.1 cases per 100,000, over the 4-year period from 2015 to 2019 nationally[2]. In 2020, Neisseria gonorrhoeae (NG) and primary/secondary syphilis rates increased by 5.7% and 6.8% respectively compared with 2019; chlamydia rates decreased in 2020 which may be confounded by less chlamydia screening as STI services became limited during the pandemic.[3] Extragenital STIs are often asymptomatic and may go undetected without screening. Pharyngeal gonorrhea may be more difficult to treat partly because of less good penetration of antibiotics into pharyngeal tissues compared to genital/rectal sites [4, 5]. Rectal STI dramatically increases the risk for HIV acquisition for both men and women[6] and, therefore, warrants more attention and assessment. In studies among men who have sex with men (MSM), extragenital chlamydia and gonorrhea can persist for many months and are likely to serve as a reservoir of infection[7]. The 2021 CDC STI treatment guidelines[8] recommend routine extragenital (rectum, pharynx) STI testing for MSM[9, 10], and for people with HIV, and in women in certain scenarios. However, among younger-age women and other heterosexual populations, extragenital STI testing has been reported to identify cases that would have been missed with genital-only testing[11–14].

The “IWantTheKit (IWTK)” program is a free online public health platform for home specimen self-collection and mail-in for combined Chlamydia trachomatis (CT)/NG testing by nucleic acid amplification [15]. Residents of Maryland, Alaska, Arizona, and most recently Oklahoma, can order swabs based on their perceived risks and site of exposure including genital (penile-meatal and vaginal), rectal, and pharyngeal samples. Since IWTK specimen collection is done in the privacy of one’s home and free of charge, users may be better able to request the site-specific testing they need than with traditional, provider-initiated testing. The COVID pandemic and the reduced capacity or closure of STI clinics led to increased referrals to IWTK; the number of IWTK users and the number of monthly orders increased 543% since April 2020 and have attracted users in populations with higher rates of STIs (younger, MSM, etc.) which we refer to as priority populations[16, 17].

We sought to assess the diagnostic value of extragenital testing in addition to genital testing; to evaluate the additional value of extragenital testing in rectal and in pharyngeal samples by gender, age group as well as risk quiz score group; and to determine factors associated with extragenital test positivity by organism.

Methods

Population

From August 20^th, 2013, to January 31^st, 2022, 13,748 unique users from Maryland, Washington DC (historical, ended in February 2021), Alaska, and Arizona (newly added since February 2021) ordered at least once from IWTK. Of those, 2,813 (20.5%) users placed more than one order over the 9-year period. Overall, 7,432 (54.1%) IWTK users returned their home-collected swabs for testing. Users who never had a positive STI test during the analysis period were considered STI negative. There were no rectal CT or NG positive and only 3 pharyngeal CT or NG positive tests among users whose gender identities are transgender men, transgender women, non-binary, gender queers and other, thus we limited our main analysis to users who self-identified as men or women. Among users who returned their swabs, after excluding 139 users of other gender identities, and 4 users who did not report age, 7,289 users had at least one STI test result available during the analysis period. Overall, 3,407 (24.8% out of 13,748) users requested both genital and extragenital tests and were included in the main analysis (Figure 1). The 3,407 users who had paired genital and extragenital tests (pharyngeal, rectal, or both) results were categorized into four groups in Supplemental figure 1: 1) No STI (negative at all sites);2) positive genital, negative extragenital; 3) positive extragenital, negative genital; and 4) genital and extragenital positive. Category 3) demonstrated the added value of extragenital testing.

Figure 1.

Consort diagram of the study

Risk quiz score

Sexual “risk” refers to behaviors that are most associated with higher rates of STI and include condomless sex and sex with multiple sex partners for example. The sexual risk quiz is a validated 6-question survey of age and STI-related behavior completed by users when placing an IWTK order. Detailed content of the risk quiz questions is described elsewhere. [18] Briefly, the risk quiz covered questions of age range, sex partners, STI history and condom use, and score was calculated by adding values of assigned points to each response with a range of 0–10. Since February 2021, IWTK users were required to finish the risk quiz before they can order an STI test. In two previous studies[18, 19], risk quiz scores were categorized into three groups depending on gender. In women, 0–4 was designated as “low risk”, 5–7 as “medium risk” and 8–10 as “high risk”; in men, 0–2 was designated as “low risk”, 3–6 as “medium risk” and 7–10 as “high risk”. STI prevalence significantly increased with higher risk score category in both men and women. Here we used the same risk score classification to assess its association with extragenital STIs.

Analyses

Among 13,748 users who had ever placed at least one STI testing order during the analysis period, we assessed types of samples ordered by men, women and other self-identified genders (a composite category including transgender men, transgender women, gender queer, non-binary and other). The ordering pattern (in table 1) was assessed for “other” gender category, however due to limited number of positive tests in this group, results were not assessed. Among 7,289 users who had returned samples and had at least one STI test result available, the proportion positive for chlamydia and gonorrhea in genital, pharyngeal and rectal samples were stratified by male and female gender and age group. Differences in proportion of positive test by age group in table 2, and demographic distributions in table 3 were tested using chi-squared test or Fisher’s exact test when the expected frequency is less than 5 in some cells. Among 3,407 users who had paired genital-extragenital test results, the numbers and percentages in the four test result groups were summarized for chlamydia and gonorrhea respectively by gender, and then further stratified by age groups and risk quiz score groups. A backward selection was used to determine the final model in which race group was dropped for best fit. The odds ratios of different factors, including gender, age group and risk score groups, associated with having a positive CT and NG extragenital test were estimated using a logistic regression model.

$$\begin{gathered} \log \left(oddsofextragentialpositive\right)={\beta }_0+{\beta }_1·\left(gender\right)+{\beta }_2·\left(agegroup\right)+{\beta }_3· \\ \left(gender\mathrm{*}agegroup\right)+{\beta }_4·\left(riskquizscoregroup\right)+\epsilon \end{gathered}$$

Table 1.

IWTK sample type ordered by gender.

	Women	Men	Transmen/transwomen/non-binary/gender queer/other
Sample type ordered:	N=7461	N=6005	N=282
Genital only	4087 (54.8%)	2907 (48.4%)	58 (20.6%)
Genital + pharyngeal	1530 (20.5%)	1025 (17.1%)	73 (25.9%)
Genital + rectal	1284 (17.2%)	1269 (21.1%)	75 (26.6%)
Genital + pharyngeal + rectal	542 (7.3%)	655 (10.9%)	68 (24.1%)
Pharyngeal + rectal	1 (<0.1%)	56 (0.9%)	7 (2.5%)
Pharyngeal only	14 (0.2%)	32 (0.5%)	1 (0.4%)
Rectal only	3 (<0.1%)	61 (1.0%)	0 (0.0%)

Table 2.

Positivity of chlamydia and gonorrhea by anatomical site, gender, and age group.

	Women					Men
	[ALL]	< 25 y/o	25–34 y/o	≥ 35 y/o	P-value	[ALL]	< 25 y/o	25–34 y/o	≥ 35 y/o	P-value
	N=3982	N=1485	N=1739	N=758		N=3307	N=893	N=1536	N=878
Genital chlamydia	328 (8.2%)	190 (12.8%)	114 (6.6%)	24 (3.2%)	<0.001	240 (7.3%)	90 (10.1%)	123 (8.0%)	27 (3.1%)	<0.001
Pharyngeal chlamydia1	16 (1.4%)	10 (2.2%)	6 (1.1%)	0 (0.0%)	0.049	4 (0.4%)	2 (0.7%)	2 (0.4%)	0 (0.0%)	0.563
Rectal chlamydia	47 (5.3%)	28 (10.0%)	15 (3.9%)	4 (1.8%)	<0.001	74 (6.6%)	19 (6.7%)	33 (6.3%)	22 (7.1%)	0.920
Genital gonorrhea	30 (0.8%)	17 (1.1%)	10 (0.6%)	3 (0.4%)	0.079	36 (1.1%)	10 (1.1%)	19 (1.2%)	7 (0.8%)	0.602
Pharyngeal gonorrhea	7 (0.6%)	5 (1.1%)	2 (0.4%)	0 (0.0%)	0.228	15 (1.5%)	1 (0.4%)	10 (2.0%)	4 (1.7%)	0.192
Rectal gonorrhea	8 (0.9%)	4 (1.4%)	3 (0.8%)	1 (0.4%)	0.602	32 (2.9%)	10 (3.5%)	12 (2.3%)	10 (3.2%)	0.567

¹Percentage represents number of positive cases over number of tests completed for that sample type.

Table 3.

Demographics among 3,407 users who had paired genital and extragenital test results.

	Women	Men	P-value	N
	N=1780	N=1627
Age group:			<0.001	3407
< 25 y/o	638 (35.8%)	432 (26.6%)
25–34 y/o	775 (43.5%)	779 (47.9%)
≥35 y/o	367 (20.6%)	416 (25.6%)
Region:			0.001	3407
Alaska	276 (15.5%)	182 (11.2%)
Arizona	13 (0.7%)	11 (0.7%)
Maryland and DC	1491 (83.8%)	1434 (88.1%)
Hispanic	92 (10.7%)	84 (12.2%)	0.412	1551
Risk quiz score group:			<0.001	3034
Low risk	472 (30.3%)	113 (7.7%)
Medium risk	946 (60.8%)	847 (57.3%)
High risk	139 (8.9%)	517 (35.0%)
Race:			<0.001	3407
Alaska Native/American Native	95 (5.3%)	45 (2.8%)
Asian/Pacific Islander	94 (5.3%)	114 (7.0%)
Black	756 (42.5%)	575 (35.3%)
Multiracial	144 (8.1%)	150 (9.2%)
Unknown	62 (3.5%)	52 (3.2%)
White	629 (35.3%)	691 (42.5%)

Where extragenital positive was defined as belonging to either group 3) “positive extragenital, negative genital” or group 4) “genital and extragenital positive”. Odds ratio (OR) and 95% confidence interval (CI) were reported. All analyses were performed using R version 4.2.1. [20]

Results

Assessment of sample type ordered (N=13,748)

Among 13,748 users who ordered tests, 54.8% of women ordered genital tests only, 20.5%, 17.2%, and 7.3% ordered both vaginal and pharyngeal, vaginal and rectal, and all three sites, respectively; 48.4% of men ordered genital tests only, 17.1%, 21.1% and 10.9% ordered penile and rectal, penile and pharyngeal, and triple sites, respectively. Overall, more men requested genital plus at least one extragenital test than women. However, a high proportion of women ordered pharyngeal testing in addition to genital testing compared to men (37.7% vs 28.0%), and a higher proportion of men ordered rectal testing in addition to genital testing compared to women (32.0% vs 24.5%). The highest proportion (24.1%) of triple-site tests (genital, pharyngeal, and rectal) were ordered by users who self-identified as other self-identified genders. (Table 1)

STIs positivity by male-female sex and age group (N=7,289)

Among 7,289 users in the analysis, 328 (8.2%) women had a positive vaginal chlamydia result, while 240 (7.3%) men had a positive penile chlamydia result. Women also had higher positivity of pharyngeal chlamydia than men (1.4% vs 0.4%), respectively. Positivity for genital chlamydia was significantly higher in age group < 25 y/o, compared to those who are 25 years and older, regardless of gender (p<0.001). The same trend where positivity decreased with age was also observed in pharyngeal (p=0.049) and rectal (p<0.001) chlamydia among women. Among men, the proportion with rectal chlamydia remained at comparable levels across age categories. Overall, positivity for gonorrhea was lower than chlamydia, and more common in men than women at all sites except in women aged <25 years where pharyngeal infection was more common than in men in the same age group. Most extragenital gonorrhea cases were detected in male rectal samples, 32 (2.9%) of 3308 rectal samples tested among men. There was no significant difference for gonorrhea positivity in any anatomical site by age group, regardless of gender. Women aged <25 years had higher or equal positivity for CT/NG than men in the same age category for all sites except for rectal NG. (Table 2)

Additionality of extragenital testing (N=3,407)

The demographic characteristics of the 3,407 users with available contemporaneous paired genital and extragenital test results are shown in Table 3. More women than men were <25 y/o, there was a higher proportion of black than white women; this was reversed in men. Significantly more men were in the “high risk” group than women, 35.0% of men compared to only 8.9% of women. Table 4 summarizes the numbers and proportions of four result categories in rectal and in pharyngeal stratified by gender and age group. Rectal testing identified 62 (5.9%) additional CT and 29 (2.7%) additional NG cases that were negative in genital in men and an additional 12 (1.3%) NG and 2 (0.2%) CT in women. There were 37 (4.1%) women who had chlamydia co-infection in genital and rectal samples. Similarly, pharyngeal testing identified 30 (3.1%) additional CT and 18 (1.8%) additional NG in men. In women, pharyngeal testing identified 12 (1.0%) CT, and 3 (0.3%) NG that would have been missed by genital-only sampling; 16 (1.4%) had co-infection of CT in genital and pharyngeal samples. Further categorized by age group, 23 of 37 (62%) genital-rectal CT co-infections and 12 out of 16 (75%) genital-pharyngeal chlamydia co-infections were detected in women <25 y/o. Similarly, the highest pharyngeal chlamydia positivity was also observed in <25 y/o women, but to a lesser extent. However, there was no extragenital positivity difference by age group observed in men.

Table 4.

Positivity of paired genital and extragenital testing by gender and age group.

	Women					Men
	[ALL]	< 25 y/o	25–34 y/o	≥ 35 y/o	N	[ALL]	< 25 y/o	25–34 y/o	≥ 35 y/o	N
	N=1177	N=446	N=523	N=228		N=1057	N=273	N=490	N=295
	Genital & Rectal Chlamydia:				893	Genital & Rectal Chlamydia:				1057
Gen(−), Rectal(−)	828 (92.7%)	244 (87.5%)	363 (94.0%)	221 (96.9%)		939 (88.8%)	235 (86.1%)	438 (89.6%)	266 (90.2%)
Gen(−), Rectal(+)	12 (1.3%)	5 (1.8%)	5 (1.3%)	2 (0.9%)		62 (5.9%)	16 (5.9%)	28 (5.7%)	18 (6.1%)
Gen(+), Rectal(−)	16 (1.8%)	7 (2.5%)	6 (1.6%)	3 (1.3%)		49 (4.6%)	21 (7.7%)	19 (3.9%)	9 (3.1%)
Gen(+), Rectal(+)	37 (4.1%)	23 (8.2%)	12 (3.1%)	2 (0.9%)		7 (0.7%)	1 (0.4%)	4 (0.8%)	2 (0.7%)
	Genital & Rectal Gonorrhea:				893	Genital & Rectal Gonorrhea:				1057
Gen(−), Rectal(−)	881 (98.7%)	273 (97.8%)	382 (99.0%)	226 (99.1%)		1021 (96.6%)	264 (96.7%)	474 (96.9%)	283 (95.9%)
Gen(−), Rectal(+)	2 (0.2%)	1 (0.4%)	0 (0.0%)	1 (0.4%)		29 (2.7%)	7 (2.6%)	13 (2.7%)	9 (3.1%)
Gen(+), Rectal(−)	3 (0.3%)	1 (0.4%)	1 (0.3%)	1 (0.4%)		5 (0.5%)	1 (0.4%)	2 (0.4%)	2 (0.7%)
Gen(+), Rectal(+)	7 (0.8%)	4 (1.4%)	3 (0.8%)	0 (0.0%)		2 (0.2%)	1 (0.4%)	0 (0.0%)	1 (0.3%)
	Genital & Pharyngeal Chlamydia:				1177	Genital & Pharyngeal oat Chlamydia:				976
Gen(−), Phar(−)	1091 (92.7%)	389 (87.2%)	497 (95.0%)	205 (98.6%)		889 (91.1%)	229 (88.4%)	450 (91.8%)	210 (92.5%)
Gen(−),Phar (+)	12 (1.0%)	7 (1.6%)	5 (1.0%)	0 (0.0%)		30 (3.1%)	9 (3.5%)	12 (2.4%)	9 (4.0%)
Gen(+),Phar (−)	58 (4.9%)	38 (8.5%)	17 (3.3%)	3 (1.4%)		55 (5.6%)	20 (7.7%)	28 (5.7%)	7 (3.1%)
Gen(+),Phar (+)	16 (1.4%)	12 (2.7%)	4 (0.8%)	0 (0.0%)		2 (0.2%)	1 (0.4%)	0 (0.0%)	1 (0.4%)
	Genital & Pharyngeal Gonorrhea:				1177	Genital & Pharyngeal Gonorrhea:				976
Gen(−),Phar (−)	1166 (99.1%)	440 (98.7%)	520 (99.4%)	206 (99.0%)		951 (97.4%)	253 (97.7%)	476 (97.1%)	222 (97.8%)
Gen(−),Phar (+)	3 (0.3%)	2 (0.4%)	1 (0.2%)	0 (0.0%)		18 (1.8%)	4 (1.5%)	9 (1.8%)	5 (2.2%)
Gen(+),Phar (−)	4 (0.3%)	1 (0.2%)	1 (0.2%)	2 (1.0%)		4 (0.4%)	1 (0.4%)	3 (0.6%)	0 (0.0%)
Gen(+),Phar (+)	4 (0.3%)	3 (0.7%)	1 (0.2%)	0 (0.0%)		3 (0.3%)	1 (0.4%)	2 (0.4%)	0 (0.0%)

Factors associated with extragenital positivity

Women age < 25 y/o had 4-fold increased odds of extragenital chlamydia compared to age ≥ 35 (p=0.010). Among men, however, age was not significantly associated with odds of extragenital CT/NG. For both men and women, being in the “high risk” group was independently associated with extragenital chlamydia and gonorrhea compared to being in the “low risk” group. (Table 5)

Table 5.

Logistic regression results of extragenital chlamydia and gonorrhea on age group, gender, and risk quiz score group.

	Extragenital Chlamydia			Extragenital Gonorrhea
	OR1	95% CI	P-value2	OR	95% CI	P-value
Age group (ref: ≥35 y/o)
Female: <25 y/o	4.0	1.4–11.4	0.010*	3.1	0.4–25.3	0.295
Female: 25–34 y/o	2.0	0.7–6.1	0.201	1.8	0.2–15.9	0.614
Male: <25 y/o	0.9	0.5–1.8	0.813	0.7	0.3–1.6	0.357
Male: 25–34 y/o	0.8	0.5–1.5	0.550	0.7	0.3–1.5	0.350
Risk quiz score group (ref: Low risk)
Medium risk	1.4	0.7–2.5	0.321	2.3	0.5–10.3	0.263
High risk	2.6	1.3–5.0	0.005*	8.5	1.9–37.7	0.005*

¹OR – Adjusted odds ratio from logistic regression model.

²Asterisk (*) indicates p-value for corresponding odds ratio was significant at 0.05 level.

Discussion

We assessed the ordering patterns, prevalence of CT/NG, and diagnostic value of extragenital testing using data routinely collected from users requesting STI testing through the online IWTK platform. We first assessed users’ order patterns and found that a higher proportion of men requested genital plus an extragenital test than women based on perceived exposure. In another study, 35.9% of women aged 18–44 reported anal sexual behavior[21], whereas 24.5% of women in our analysis requested rectal sample testing. This is in stark contrast with other studies in non-STI clinic attendees, in a study of 7.9 million women from all US states and DC who had CT/NG testing at a large commercial laboratory between 2012 and 2015, 0.1% and 0.5% had rectal and pharyngeal tests requested by their providers respectively[21]. The authors conclude that most women who have anal sex are not being tested for rectal CT or NG. This suggests online-ordered, self-collection scheme of STI testing can successfully target populations who had extragenital-site sexual exposure and potentially allows for better detection of STIs. This represents an important opportunity for site-specific testing based on an individual’s perceived exposure and overcome barriers introduced by clinician-directed testing including ascertainment and social desirability biases.

STI data available from the CDC are limited because infection by anatomical site is not required to be reported to CDC, therefore national data on extragenital CT/NG are lacking. However, extragenital CT/NG is thought to contribute to community spread of infection, and a may be a risk factor for serious pathology such as disseminated gonococcal infection. [22] More data are required to better understand the full health implications of extragenital CT/NG.

We evaluated overall CT/NG positivity by anatomical sites and found that genital CT differs by age in both male and female gender; extragenital CT only differed by age in women. Young women IWTK users had concerningly high CT/NG positivity in all sites except rectal NG suggesting that women under the age of 25 should consider extragenital STI screening. Our data suggest that young women with extragenital exposure could become the next priority population for consideration of screening. Recent reports of extragenital STI prevalence from Chan et al[14] (2.0–77.3% for rectal chlamydia (median 8.7%)) and Andersson et al [23](rectal CT 9.1% in women) also support our findings of high rectal CT prevalence in women. Those finding are mostly in clinical settings, a more thorough evaluation of rectal STI in young women outside of clinical settings may be warranted because demographics and behavioral associations may be different in women using home-collected samples. Rectal STI positivity is a marker of high risk sexual behaviors and the inflammation associated with rectal infection is associated with an increased risk of HIV acquisition.[24] While experts debate if rectal CT represents autoinoculation from vaginal secretions or infection by receptive anal sex [17], the high proportion of women requesting rectal testing in IWTK users suggest that they at least perceived themselves to have a rectal exposure. A previous study found that 59% of women with rectal CT denied anal intercourse[23], and another study reported no difference in positivity whether receptive anal intercourse was reported,[25] suggesting women are still at risk of rectal CT regardless of rectal sex exposure.

Additionally, we assessed the additional diagnostic yield of extragenital testing and found that in the IWTK users, extragenital testing identified a substantial number of STIs that would have been missed by only offering genital testing. Furthermore, we found dual positivity, in genital and extragenital sites, for both CT and NG was higher in women than men.

In the logistic regression model, odds of extragenital CT infection significantly increased among young women and odds of extragenital CT and NG also significantly increased among people with high risk quiz scores. These findings suggest that extragenital testing should be recommended for young women (<25 years old) and IWTK users with a high score in the sexual risk quiz. It is possible that IWTK users were only diagnosed because online testing was low barrier or convenient, and likely that the additional cases identified by extragenital testing would have been missed because generally providers may fail to offer testing by exposure site. Extragenital STIs have been frequently evaluated among MSM, in women, a previous analysis yielded similar results as this analysis, that cases of CT/NG would have been missed with genital-only testing, especially among young women [26–29].

Extragenital, specifically rectal, STIs increase the risk of HIV acquisition, because extragenital STIs are most common in MSM, screening is prioritized in these populations. However, STIs also have other detrimental reproductive health effects such as ectopic pregnancy, especially in young women. Extragenital STIs are predominantly asymptomatic and can persist for long periods of time with increased opportunity for transmission. Pharyngeal commensal Neisseria species play a role in NG antimicrobial resistance, extragenital NG risk for disseminated gonococcal infection, extragenital CT may be a risk for immune-mediated effects of CT such as sexually associated reactive arthritis.

Each additional STI test comes at an additional cost. The apparent benefits of testing for and treating extragenital STIs have to be weighed against the additional cost. The higher costs could be a major barrier to widespread adoption and implementation of extragenital testing. Although we did not perform a cost-effectiveness evaluation of extragenital testing, a previous study evaluated the cost effectiveness and found that performing triple site (genital, rectal, and pharyngeal) testing is the most cost-effective model and decrease prevalence and incidence from 3 years onward. Another study of California young adults found that two site (genital and rectal) testing could identify greatest proportion of infected individuals.[27] More cost-effectiveness studies in non-clinical settings are also needed to better evaluate cost-benefit ratio of multiple-site testing.

Strengths of this analysis include large sample size, similar proportion of men and women, over a long time period, black women were particularly well represented. This was, of note, the first study to evaluate associations of behavioral risk, as measured by an online sexual health risk score, and extragenital STIs. It gives insight on values and performance of self-collecting specimens as a mean of STI screening in the general population in Baltimore city, Washington DC and Alaska. Our analyses had limitations; pharyngeal testing was paused from 9/1/2020 to 4/6/2021 due to COVID-10-associated supply limitations therefore we likely underestimated pharyngeal infections. Symptoms, or sex of sex partner data were not collected by IWTK during the analysis period, so we were unable to assess the relationship between these and extragenital testing or positivity. The analysis was limited to users of an online testing platform, therefor results cannot be generalized to other populations.

Conclusion

Based on these data, women younger than 25 years old could potentially be the next priority population in whom to recommend extragenital testing. Individuals at high risk, including those with high sexual health risk scores, regardless of gender or age should consider extragenital testing.