Access to this full-text is provided by Springer Nature.
Content available from Trials
This content is subject to copyright. Terms and conditions apply.
Ybarraetal. Trials (2025) 26:9
https://doi.org/10.1186/s13063-024-08540-9
STUDY PROTOCOL Open Access
© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0
International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or
parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To
view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Trials
Sexual Health Advocacy forGuys
(SHAG): arandomized trial oftheimpact
ofatext-messaging program onHIV incidence
andSTI testing amonganational sample
ofsexual minority cisgenderadolescent
andyoung adult men
M. L. Ybarra1* , D. J. Feaster2, R. Garofalo3 and S. Bull4
Abstract
Background Disparities in sexually transmitted infections (STI) including human immunodeficiency virus (HIV)
among sexual minority boys and young men are substantial. Effective HIV and STI prevention programs that include
access to pre-exposure prophylaxis (PrEP) medication do not consistently include younger sexual minority men. Text-
messaging programs for HIV prevention have been associated with increases in HIV testing among sexual minority
adolescent boys, but these programs have not incorporated a focus on PrEP or STIs beyond HIV.
Methods We will conduct a two-arm randomized trial with 1:1 allocation comparing the superiority of text messag-
ing-based intervention focused on HIV and STI prevention to a generic HIV education program with content focused
on promoting a “healthy lifestyle” (e.g., self-esteem). Outcomes include testing for HIV and other STIs, increasing PrEP
and PEP use, and HIV incidence. Generalized linear models will be used to estimate treatment effects on primary study
outcomes, with longitudinal models (estimated based on Generalized Estimating Equations) specified to examine
effects over time. Mediation will be assessed based on a product of coefficients approach with bootstrapped standard
errors.
Discussion This is the first randomized controlled trial (RCT) with a national sample of cisgender sexual minority
adolescent boys and young men 13-22 years of age exploring the efficacy of a text messaging-based intervention
in increasing HIV and STI testing, and PReP and PEP use. Findings will inform the scalability of text messaging pro-
grams for sexual health promotion and at-home STI testing, and will demonstrate impacts of a behavioral health
intervention on HIV incidence.
Trial registration ClinicalTrials.gov NCT06 230367. Date of registration: 1/29/2024.
Keywords Text messaging intervention, Sexual health, Sexual minority health, HIV testing, STI testing, Home-testing,
HIV PReP, HIV PEP
*Correspondence:
M. L. Ybarra
michele@innovativepublichealth.org
Full list of author information is available at the end of the article
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 2 of 21
Ybarraetal. Trials (2025) 26:9
Introduction
Background andrationale
Sexual minority cisgender adolescent boys and young
menface disparate risk for HIV acquisition: More than
nine in ten new HIV infections among boys and young
men in the USA are through “male-to-male” sexual con-
tact [1, 2]. Disparities are even starker for African Ameri-
can/Black and Hispanic sexual minority youth, who
account for 51% and 25% of new HIV infections, respec-
tively [3]. Youth living in southern states [1] and in rural
settings [4, 5] also have higher rates of HIV. While the rea-
sons are not entirely clear, it may in part be due to reduced
access to HIV counseling and the availability of preventive
services, as well as less accepting attitudes towards sexual
minority people. HIV testing and counseling as well as
pre-exposure prophylaxis (PrEP), which can reduce one’s
risk of HIV up to 99% [6–11], are critical components of
any comprehensive HIV prevention initiative.
In the face of all-time high STI rates across the coun-
try [12], it is concerning that youth ages 15–24 account
for half of all new infections [12, 13], invigorating calls for
increased prevention focus on STIs [12]. STI testing—
including oral and anal tests [14]—is important because
people who have non-HIV STIs are more likely to con-
tract HIV [15–22]. is is partly due to concomitant risks
(e.g., unprotected sex), but also because a sore or inflam-
mation caused by an STI provides a pathway for HIV.
e wide adoption of text messaging provides novel
opportunities for HIV prevention interventions where
youth “are” across socio-demographically different
groups [23], and overcomes structural challenges of tra-
ditional prevention initiatives. Moreover, being able to
access sensitive content when and where one chooses
facilitates safe spaces for youth to engage with the con-
tent, which is important for those who are not “out” to
family and friends. Importantly too, reviews suggest that
programs delivered via text messaging can affect complex
behavior change, including HIV testing among sexual
minority adolescent boys [24–29].
e Guy to Guy program (G2G) was the first compre-
hensive HIV prevention program delivered via text mes-
sage to a national sample of 14–18-year-old cisgender
boys who identified as gay, bisexual, and/or queer. G2G
was tested against an attention-matched “healthy life-
style” control focused on topics such as self-esteem. G2G
and the healthy lifestyle control programs sent between
5 and 10 messages per day to participants over a 5-week
period. At 6 weeks post intervention, participants in
each program received booster messages for a week. At
90days postintervention, there were no significant dif-
ferences in either sex acts not protected by condoms or
abstinence between groups. Among participants who
were sexually active at baseline, intervention participants
reported a threefold increase in HIV testing compared
to control participants (adjusted odds ratio = 3.42,
P = 0.001). ey were less likely than control youth to be
abstinent (adjusted odds ratio = 0.48, P= 0.05) [29].
G2G offers an example of a promising program that
wasimplemented and testedat the national level, afford-
ing substantial reach and impact beyond traditional
face-to-face initiatives for gay, bisexual, and other sexual
minority cisgender adolescent boys and young men. But
G2G did not incorporate a focus on PrEP, with critical
information on what it is, how to access it, and when it
may be indicated for use. Additionally, few studies are
powered to a degree that supports analyses of HIV inci-
dence. Given widespread availability of home testing for
HIV, we now have the possibility to afford participants
the privacy and convenience of a home HIV test. is
has the added benefit of being able to photo-validate
self-report of HIV testing while also documenting HIV
incidence in a hard-to-reach audience at risk. Finally,
while G2G impacts on HIV testing are important, it also
is important to address common concomitant sexually
transmitted infections.
In this paper, we present details on the protocol for a
randomized controlled trial of Sexual Health Advocacy
for Guys (SHAG), designed to test the impact of an inter-
vention on HIV incidence and STI testing. SHAG is a text
messaging-basedintervention that builds on and adapts
elements from multiple previously evaluated inter-
ventions, including G2G, Girl2Girl, a text messaging-
basedintervention focused on pregnancy prevention for
lesbian and gay cisgender adolescent girls [30, 31] and
In is ToGether, a text-messaging program focused on
HIV prevention for Ugandan young adults [32]. SHAG is
delivered via text message to sexual minority csigender
adolescent boys and young men.
Objectives
Our overall objective is to estimate the effect of SHAG
compared to a generic HIV education program among
13–22-year-old sexual minority adolescent boys and
young men across the USA. To do so, we will first adapt
content to integrate lessons learned in previous text mes-
saging-based sexual health interventions, as well as to
include information on PrEP and PEP, and add messages
that are more age-appropriate for sexually active young
adult men. Next, we will demonstrate the feasibility of
using OraQuick home tests to collect HIV test results via
photo-verificationamong this age group. Finally, we will
test the impact of SHAG on each of our primary study
outcomes (1) HIV incidence, (2) self-reported PrEP and
PEP use, (3) HIV status, and (4) number of STIs in the
past 12months.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 3 of 21
Ybarraetal. Trials (2025) 26:9
Trial design
is will be a two-arm randomized controlled trial with a
1:1 allocation ratio comparing the superiority of exposure
to SHAG versus an attention control condition.
Secondary outcomes include (1) information about
and (2) motivation for uptake of PrEP; (3) STI testing;
and (4) the impact of the intervention on mental health
indicators.
Methods: participants, interventions,
andoutcomes
e following is presented per the Standard Proto-
col Items: Recommendations for Intervention Trials
(SPIRIT) Reporting Guidelines [33].
Study setting
e SHAG intervention is designed for delivery via text
message to people across the USA who own their own
cell phone and are enrolled in unlimited text messaging
plans.
Eligibility criteria
Eligibility criteria are meant to result in a study sample
that approximates young people who would be most
likely to take part in the intervention if it were publicly
available. us, youth will:
(a) Have been assigned male sex at birth and currently
have a cisgender identity,
(b) Be aged 13–22years old,
(c) Have had anal sex in the past 12months,
(d) Be English-speaking,
(e) Exclusively own a cell phone with an unlimited text
messaging plan and intend to have the same cell
phone number for the next 6months,
(f) Have Internet access to complete online surveys,
(g) Provide informed assent for those under 18 and
consent for those 18years of age and older, includ-
ing a capacity to consent [34] and a positive self-
safety assessment [35],
(h) Willing to take an OraQuick home test to confirm
HIV negativity for youth who are 19–22 years of
age or 18years old and graduated high school. If
they agree to do the test but do not upload a photo
of their result, they will be eligible if they self-report
a negative HIVserostatus.
Youth 18 years old who have not graduated high
school and youth13–17 years of age will be given
the option to take a home-based HIV test. If they
determine that they cannot do so safely, they will be
allowed to self-report their HIVserostatus; and
(i) Not becurrently enrolled in another HIV prevention
program or know anyone already enrolled in the
SHAGRCT.
Justication forthese criteria
(1) Both our own experience and other national data show
that low-income youth are as likely as higher income youth
to have unlimited text messaging plans [36]. Only 2% of
screeners in Girl2Girl were ineligible because they did not
exclusively own a cell phone with an unlimited text mes-
saging plan. We believe text messaging increases access
to the intervention rather than reifies the impacts of the
digital divide. (2) We include all recently sexually active
youth irrespective of whether they have used condoms
because HIV incidence rates are based upon all sexual
minority boys—not just those who report recent unpro-
tected sex. (3) We include 13-year-olds to address gaps in
HIV prevention for younger audiences. (4) Most Hispanic
youth speak English; therefore, we believe that deliver-
ing the intervention in English will still be inclusive [37].
(5) Finally, we exclude friends of those already enrolled
because of potential contamination if one is randomized to
the intervention and the other to the control arm.
Gender diverse youth will be excluded because the nec-
essary tailoring of content that is appropriately gender
affirming and speaks to the unique factors impacting gen-
der diverse youth’s HIV preventive behavior is incredibly
important and also beyond the scope of the current study.
Determining co‑enrollment ofstudy participants inother
clinical trials
We will rely on self-report of current participation in
other clinical trials. In previous RCTs, we have found that
when asked, youth will tell us without hesitation if they
were referred by a friend into the study. is is because
we do nothing to suggest that they have done something
“wrong.” We anticipate similar disclosure for both ques-
tions in the current trial.
We will ask youth who indicate they are part of another
clinical trial, the name of the HIV prevention program.
We will use this information to endeavor to find informa-
tion about the trial online. If we cannot find the program
online, we will contact the youth and ask the youth to
describe the program in detail. It may be that the pro-
gram has not been registered online (e.g., in ClinicalTri-
als.gov); or that youth appraise it to be a clinical trial but
it is not (e.g., maybe they are doing a survey but it is not
an intervention). One example of an intervention that
would meet exclusion criteria is “Hey Friend,” as regis-
tered on ClinicalTrials.org (https:// clini caltr ials. gov/ ct2/
show/ NCT04 846946? recrs= ab& type= Intr& cond= Hiv&
cntry= US& gndr= Male& age= 0& draw= 2& rank=2).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 4 of 21
Ybarraetal. Trials (2025) 26:9
Who will take informed consent / assent?
We have obtained a waiver of documentation of informed
assent/consent so that a verbal or clicked “yes” will be
sufficient; a signature will not be required.
Obtaining informed consent
For those who are 18years of age and older and not in
high school, the consent will be self-administered via
an online website. is includes a self-safety assessment
with graphics to encourage participants to think through
different scenarios that could potentially place them in
danger (e.g., a partner intercepts text messages about
anal sex). ey also will complete an automated capac-
ity to consent, which queries whether the person under-
stands their voluntariness, the risks of participating, etc.
If the person indicates that they may not be safe, they will
be encouraged to not take part in the study. ose who
have questions about the study or want to confirm that it
is a “real” research study will be encouraged to reach out
to study staff via text message, email, or phone.
e self-enrollment website will be developed in Year
1 of the grant and function very similarly to the self-
enrollment website we have used for Girl2Girl, includ-
ing the automated self-safety assessment [30, 38].
Obtaining informed assent
We have obtained a waiver of parental permission for
participants under 18 because requiring parental permis-
sion could increase risk to participants who may be vic-
timized by their parents because of the need to disclose
their sexual minority status. A waiver also is necessary to
avoid fatal sampling bias that would occur by only includ-
ing youth who are out to their parents.
e research staff, all of whom are trained in Human
Subjects Protections, will discuss over the telephone
assent information with the candidates who are 17years
of age and younger, or those who are 18years old and still
in high school. e participant will be asked to complete
a brief capacity to assent assessment, which will measure
his ability to understand the potential risks of partici-
pation [39]. Specifically, participants will be quizzed to
assess their capacity to understand, appreciate, reason
with, and express a choice about participation using a
modified version of the Evaluation to Consent Form [34,
39, 40]. Modifications involve the specific risks that could
result from participation (e.g., “If someone sees one of
the project texts, they may ask me about my sexual iden-
tity”). Participants who do not demonstrate a capacity to
assent will be ineligible.
Given the high rates of interpersonal victimization that
sexual minority youth report [41], we believe that chil-
dren who are eligible for this study will be well equipped
to self-assess their safety, as they have to do so every day
of their lives. Study staff will complete a self-safety assess-
ment with the potential participant that includes dis-
cussion of different possible scenarios and asking them
whether they feel safe in each situation. is conversation
will include concrete examples, such as their caregivers
monitoring their cell phone use and text message con-
tent. If the participant is hesitant at all, he will be advised
not to participate. While this protocol will likely result
in fewer participants assenting, it will result in a safer
cohort. e sample also will be more reflective of the
actual end user. While it may be uncommon to empower
youth to make their own safety decisions, this self-safety
assessment was used successfully in Guy2Guy, Girl2Girl,
and In is toGether.
All participants will be given the opportunity to ask
any questions. A link to the assent / consent form will be
sent via email for his later reference, should he request it.
[Some youthmay not desire it be sent for safety reasons.]
Phone numbers for the Principal Investigator (PI) and
theInstitutional Review Board (IRB) will be listed in the
assent/consent forms and on the intervention website, in
case a participant has a question or would like to discuss
the study or any concerns.
Additional consent provisions forcollection anduse
ofparticipant data andbiologic specimens
Home‑based HIV tests
During enrollment, older youth will be asked to think
how to safely receive the mailed home-based HIV test.
Younger youth who chose to receive a home-based test
will be asked to think about the same.
If it is safer for youth to have the package mailed to
an address other than their home (e.g., because a parent
may question what is inside), we will send it to another
address that they specify. We will send the USPS track-
ing information to participants so that they can opt-in
to real-time tracking of when the package is going to
be delivered and are empowered to redirect the pack-
age should they need or want to. If young adults are not
able to identify during enrollmenta place where we can
safelymail tothem the package, we may offer the option
of mailing the test to an Amazon Hub Locker. In these
instances, the test will be purchased and mailed by Ama-
zon. Younger participants (i.e., those who are 13-17 years
of age or 18 years of age and still in high school) whodo
not think they can receive or do the home-based HIV test
safely may opt out.
Interventions
Explanation forchoice ofcomparators
Text messaging interventions and other interventions
that rely on technology solutions design attention con-
trol comparators to ensure that observed effects can be
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 5 of 21
Ybarraetal. Trials (2025) 26:9
appropriately attributed to intervention content rather
than to the benefits of receiving messages only.
e comparison group in this trial will receive a similar
number of text messages for the same number of days as
the intervention group. e content for the comparison
group will focus on generic HIV prevention education
and other healthy lifestyle topics (e.g., moving your body,
self-esteem). Dyadic messaging features included in the
intervention to increase program engagement are not
included in the control content.
Intervention description
Intervention messages will be conversational in tone
and build upon each other daily and weekly throughout
the intervention. “Core” content will be delivered across
~8weeks. Between 8 and 15 program messages will be
sent each day. Although this may seem like a lot, half of
teens send 60 text messages a day [42], which is likely
why we have found this intensity to be acceptable to
young people [43, 44].
As shown in Table1 below, messages targeting moti-
vation to engage in HIV preventive behaviors discuss
the importance of HIV testing and not assuming a part-
ner’s sero-status, and normalize the idea that condom
use is a loving way to show that you care about your and
your partner’s health. Behavioral skills include messages
abouthow to talk to a healthcare provider about PrEP
and PEP, and how to talk to one’s partner about using
condoms. Content will discuss the benefit that PrEP can
have if having condomless sex is one’s current reality,
and how PEP can be used if a single unprotected sex act
happens.
We also will link to brief videos where visual informa-
tion is critical to achieving the learning objective. For
example, messages that discuss how to use a condom
are complemented with embedded links to interactive
demonstrations.
Criteria fordiscontinuing ormodifying allocated
interventions
ere are no specific criteria to modify the control or
intervention arm content, but we do recognize that exter-
nal events may warrant consideration of new content or
adaptation of existing content. For example, during the
COVID-19 pandemic, we made changes to Girl2Girl
content that seemed to be exacerbating mental health
issues for some who were particularly stressed by the
pandemic. Although we do not anticipate something
similar, it may arise. In this case, we will document any
changes made to content and work to maximize fidelity
throughout the trial.
Participants can self-select to discontinue their partici-
pation at any time; in these instances, data collected up to
the time of discontinuation will be retained for analysis.
Strategies toimprove adherence totheintervention
Personalizing the content increases the self-relevance
of material, thereby improving the likelihood that the
information will be understood, remembered, and pro-
duce behavior change [45–47]. Examples from previous
research demonstrate that content can be written for dif-
ferent “paths” that present the same concepts but in ways
that are more relevant to specific subpopulations (e.g.,
those who are having sex with guys versus those who are
having sex with guys and women). We will explore the
possibility of creating paths to tailor content, for exam-
ple by racial/ethnic identity and urban/ rural settings.
Youth will also have opportunities to tailor the program
to their daily schedule. ey can determine when the
messages start and end each day, which ensures that they
will be sent at appropriate times (e.g., after band practice)
and therefore will be more likely read. Participants also
will determine the intensity of their messaging: A longer
window of messaging means that the messages will be
more spaced out across the day.
Table 1 Example SHAG messages
HIV Information Lube also reduces the chances that the fragile skin around the anus and in the rectum will tear. It also keeps the condom
from breaking (see how I worked that in there? 😊)
PrEP is a pill you take every day or a shot you get every 2 months. It reduces your risk of HIV by 99% (for real) when you take it
the right way.
HIV Motivation I know it may feel like the 2000s to talk about HIV, but the truth is: Among people who are living with HIV, 2 in 3 are guys who are
into guys. This is real. It affects all of us.
HIV Behavioral Skills Maybe trusting doctors is hard – not all doctors are LGBT+ friendly. Here’s a website with doctors who have experience working
with LGBT+ folks: lgbtqhealthcaredirectory.org. Finding a doctor now might help when you need one in the future.
You might tell your partner that you’ve learned stuff in SHAG - like how great PrEP is at preventing HIV, and condoms are at stop-
ping STIs; and that testing every 3 months is important just to be safe. You want to follow this plan.
It’s perfectly ok to start doing something even after you’ve stopped – or never done it. Every time you have sex is a new chance
to make a healthy decision.
Socio-cultural factors Violence is *never* ok. If there is violence in your relationship, you might feel like you’re the only one going through this,
but remember: You are not alone.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 6 of 21
Ybarraetal. Trials (2025) 26:9
Promoting program engagement using bi-directional
messages: ere are several features that use bi-direc-
tional messaging to increase the interactivity of the inter-
vention, which in turn, we posit will promote adherence
to the intervention. ese include offering “Text Bud-
dies”, i.e., pairing participants with others in the program
that they can communicate with via text for the purpose
of practicing newly acquired communication skills [48];
“gamifying” content, a strategy demonstrated to increase
engagement with and commitment to an intervention
[49] with elements such as points and leveling up to more
challenging content; and opportunities to earn badges to
demonstrate competency of HIV preventive behavioral
skills.
e intervention will link to a screening tool that will
help youth better understand their HIV risk and to iden-
tify preventive options that best fit them.
Relevant concomitant care permitted orprohibited
duringthetrial
Potential participants are not eligible for SHAG if they
are currently enrolled in other HIV prevention programs
or if they are HIV positive at baseline. People who sero-
convert during the RCT will be encouraged to access
care and will be able to remain enrolled should they so
choose. Everyone will be encouraged to talk with a medi-
cal provider about PrEP and whether it is a good fit for
them.
Provisions forpost‑trial care
As a behavioral health study, we do not anticipate physi-
cal harm because of the study and so do not have provi-
sions for compensationor post-trial clinical care. Youth
will be provided referrals to other organizations through-
out the study should they want to continue to engage in
healthy sexuality discussionsand related care.
Outcomes
Measures foroutcomes oftheRCT
Because PrEP is an outcome, providing it as part of the
study would negate the possibility of measuring the inter-
vention’s impact on its uptake. e primary and second-
ary outcomes are as follows:
Primary outcome measure
HIV Incidence determined by home testing kit.
[Time Frame: 12-month post-intervention and imme-
diately post-intervention].
Secondary outcome measures
2. Self-reported HIV incidence
[Time Frame: Post-Intervention, 3-month post,
6-month post, 9-month post, 12-month post]
3. Proportion of participants testing for an STI
is outcome will be measured as a cumulative indica-
tor of whether the individual has tested for an STI
[Time Frame: 12-month post intervention and immedi-
ately post-intervention]
4. Proportion of participants having used PrEP/PEP
is outcome will be measured as a cumulative indica-
tor of whether the individual has used pre-exposure or
post-exposure prophylaxis for HIV
[Time Frame: 12-month post-Intervention and imme-
diately post-intervention].
Participant timeline
We have allocated at least 27 months for participant
enrollment. After enrollment, young adults, and children
who opt in, are mailed an OraQuick home-based HIV
test by Molecular Testing Labs (MTL)—see “Confirming
HIV sero-status” below.
We will reach out to those who are ineligible because
of a positive HIV test to link them to local resources
and encourage them to seek a confirmation test and
counseling, we will email candidates who are ineligible
for other reasons referrals and resources. is lag is to
reduce the likelihood that they will return to the screener
to try to enroll again but with different answers.
Conrming HIV sero‑status
We chose OraQuick over a dried blood spot, which
is more accurate, because we anticipate a higher
response rate with the home-based test. For example,
in one studywhere 15–24-year-olds were able to order
OraQuick home-based HIV tests, 65% ordered it, 75% of
whom self-reported using the test during the course of
the study [50]. In the current study, youth are being pro-
actively sent the HIV test rather than ordering it; as such,
we anticipate the 75% marker to be relevant and antici-
pate that 75% of those who are assented/consented will
complete the test a baseline.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 7 of 21
Ybarraetal. Trials (2025) 26:9
We willsend the baseline online survey to those who
are getting a home-based HIV test mailed to them on the
day the test is received (we will be able to track the test
delivery by using the USPS tracking number provided
byMTL). Participants will be told to take a picture of
the HIV test result and upload it to the survey platform.
If the photo is unreadable, we will ask the participant
to upload another picture. If the test is inconclusive, we
will ask the participant to go to a local clinic for a con-
firmatory test. If the HIVtest is positive, we will provide
resources for the participant to link them to care locally.
ose who decline to upload a photo of their test may
self-report their sero-status.
Once we have confirmed receipt of a photo of the
HIVtest and that it is negative, the participant will be
randomized and program messages will begin. e test-
ing procedure will be similar at intervention end and
12-month follow-up.
e tests will be mailed in non-descript packaging so
people handling the box will not know what it contains.
We will send the tracking information to participants if
they request to receive itso that they will know when the
package is going to be delivered and are empowered to
redirect the package should they need to review instruc-
tions that are included with the test.
Intervention length
e total intervention length will be about 5 months.
After the ~8-week “core” messaging period, participants
will enter a 12-week “latent period” during which they
will receive 2–4 messages per week that encourage them
to integrate their new HIV preventive behaviors into their
everyday lives. e intervention will end with a 1-week
“booster” session that reviews fundamental intervention
topics. Messages will highlight key messages presented
in the 8-week core content. Results from other trials sug-
gest that this latent period may be particularly important
in enacting behaviors that require a healthcare provider
(e.g., PrEP acquisition) [30, 31, 38].
Assessment timeline
Participants will complete seven surveys: At baseline,
“core” intervention end, and intervention end; and 3-,
6-, 9-, and 12-month post-intervention end. Sero-prev-
alence via home-based HIV tests will be measured at
baseline, intervention end, and 12-month post-inter-
vention end. We choose these time points to meas-
ure proximal (i.e., intervention end) and distal (i.e.,
12months post-intervention) impacts of the interven-
tion on incidence. At all other time points, youth will
self-report whether they have been tested since the last
survey, and the outcome of the test. To maximize data,
participants will be invited to complete each survey
irrespective of whether they completed the previous
survey.
Study timeline
At least 27 months will be devoted to recruitment; fol-
low-up data collection will occur through 12 months
post-intervention end for participants. e total obser-
vation period will be 17 months: 2 months for the
“core” intervention, 3months for the latent period and
review week, and 12months of follow-up. is results
in an almost 4-year field period. Data analysis will
occur in year 5.
A timeline conforming to the SPIRIT guidelines can
be found in Fig.1.
Sample size
e proposed sample size for the current study is
N = 5000 youth. We conducted power analyses to deter-
mine whether the proposed sample size is adequate to
test the intervention’s impact on reducing self-reported
HIV incidence (primary study outcome). We focused on
this outcome measure because it is the least common of
the main outcomes identified. As such, if we have suffi-
cient power to detect HIV incidence, we have sufficient
power for testing the effect of SHAG on all other primary
study outcomes.
Effect sizes for the power calculation were based on
prior work. In a study of 450 men who have sex with
men aged 16–20years old, Garofalo et al. reported the
12-month HIV incidence rate to be between 2.0 and 6.0,
with 95% confidence [51]. To be conservative, we assumed
a 30% loss to follow-up, which is considerably higher than
we have seen in previous studies by this research team.
Given that we are interested in determining whether the
intervention has a positive impact, a one-sided alpha
level was specified. Results using PASS Sample Size and
Power [52] suggest that using a log-rank test we will have
power = 0.8 to detect a hazard ratio ≤ 0.5 implying a dif-
ference between the HIV incidence rates in the treatment
(SHAG) and control if the population HIV incidence rate
is 2% or higher, with a p-value of 0.05 or less.
Recruitment
Participants will be recruited primarily through social
media and dating applications. We have had success
in prior research in recruiting large numbers of sex-
ual minority youth via social media. Instagram (IG) is
one of the most popular online platforms adolescents
use, although low-income teens are more likely to use
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 8 of 21
Ybarraetal. Trials (2025) 26:9
Facebook (FB) [53]. Because FB owns IG, ads run on both
platforms, ensuring wide visibility and reach across dif-
ferent groups of young people.
We do not ask youth to refer their friends because this
has the potential to create problems with randomiza-
tion (i.e., if one is randomized to the intervention and the
other to the control).
Given the increased vulnerability and therefore need
for protection of children under 18years of age, we have
two enrollment strategies, based upon age:
• All youth who view the IG / FB ad and want to
learn more will click on it, linking them to the pro-
ject website. ere, they read a description of the
study activity and, if interested, complete an eligi-
bility screener. Participants are not required to reg-
ister to complete the screener; this step is similar to
a contact form.
• ose who are older, defined as being 19–22years
of age or 18years of age and not in high school, and
eligible will go on to enroll themselves online by
completing an automated self-safety assessment and
agreeing to the informed consent document and
capacity to consent survey.
• ose who are younger, defined as being 13–17years
of age or 18years and in high school, will have an
enrollment telephone call with research staff to go
over the assent, capacity to assent, and self-safety
assessment together.
About 27 months will be devoted to recruitment.
As noted above, we anticipate about 75% of those who
assent/consent to take part will provide baseline HIV
testing results. If accurate, then we will need to consent/
assent 6667 young people to randomize 5000. To meet
this goal, we will need to enroll ~ 100 youth under the
age of 18 each month. If we are not receiving a sufficient
number of screeners, we will work with our research
team and young peopleto gather ideas about how to bet-
ter target younger youth.
We anticipate ~ 150 youth 18–22 years of age self-
enroll each month. [We anticipate about two-thirds
of the sample will be older because older adolescents
have sex more frequently than younger adolescents.] If
enrollment rates for older youth are lower than antici-
pated—especially because they are being asked to com-
plete an HIV test without talking with a live person on
the telephone, we will conduct phone outreach to people
Fig. 1 SHAG study timeline per the SPIRIT guidelines
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 9 of 21
Ybarraetal. Trials (2025) 26:9
who have submitted eligible screeners but have not yet
enrolled to invigorate rates if need be.
Ensuring diversity
Half of gay and bisexual men who are HIV positive are
Black or African American, as are 25% of sexual minor-
ity Latinx men [3]. As such, it is vital from a public health
perspective to ensure that racial and ethnic minority youth
are well-represented in the study. Additionally, almost one
in five people who are HIV positive live in rural areas [54].
Social media is particularly amenable to achieving sample
diversity because we can target ads on youth character-
istics. We may also reach out to social media influencers
from these harder to reach populations to promote the
study. To endeavor to have at least 50% of the sample is
Black/African American, Latinx, and/or mixed race, and
20% are living in a rural area or southern state, we will
impose diversity targets during the enrollment process.
Once we have randomized the target number of youth
from a particular “bin” (e.g., White, non-Hispanic), all sub-
sequent youth from this group will be ineligible.
Assignment ofinterventions: allocation
Sequence generation
e randomization allocation table was generated based
on a permuted block randomization procedure with
small, random-sized blocks. Randomization was strati-
fied by age (younger [defined as 13–17 years of age or
18 years of age and in high school] /older [defined as
19–22years of age or 18 years of age and not in high
school]) and sexual identity (gay/bisexual), ensuring
equal allocation across age/education x sexual identity
subgroups. e randomization sequence was generated
by the study statistician and monitored by the PI.
Concealment mechanism
Randomization assignment will be automated using a 4-,
6-, and 8-block randomly alternating design programmed
by software developers. Researchers will not have access
to the algorithm. Only the software developer and biostat-
istician will be able to view the assignments in the data.
Implementation
Youth will be enrolled sequentially within the stratifi-
cation group (or “bin”). Once a “bin” is full (e.g., White
non-Hispanic urban), no other people who have the same
profile will be eligible. Older youth will enroll themselves
online. Younger youth will be enrolled over the telephone
by research staff. Participants will be randomized after
they are enrolled and complete the baseline survey.
Assignment ofinterventions: blinding
Who will be blinded
Participants are blinded to their assigned study arm;
there is no blinding of the research staff to the interven-
tion or control arm.
Procedure forunblinding
We will communicate to participants the arm to which
they were assigned once all data collection has been
completed.
Data collection andmanagement
Plans forassessment andcollection ofoutcomes
e measures planned for collection are identified above.
Participants will upload documentation of home-based
HIV test results and self-administer surveys online.
Participants will key in their own data through online
surveys. Variables will have validity checks such that
out-of-range answers (e.g., condom use greater than the
number of times one has had sex) will be disallowed.
Data will be reviewed (blinded to treatment assignment)
on an ongoing basis to assess quality and completeness.
Plans topromote participant retention andcomplete
follow‑up
Incentives
Incentives are commensurate with those used in previous
studies of sexual and gender minority (SGM) youth [29–
31] and are purposefully nominal so that they are not
coercive. Youth in the RCT will receive graduated incen-
tives over time: $15 for the intervention end survey and
$25 for each follow-up survey except for the final survey,
which they will receive $30 for completing.
For older youth as well as younger youth who chose to
do an HIV test, they will be incentive an additional $30
for uploading the OraQuick test result at baseline, $45 at
intervention end, and $60 at 12-month post-intervention
end.
We may also offer an “early responder” incentive of an
additional $5 for those who complete the follow-up sur-
veys within the first 72hours. .
Based on prior research, some SGM youth prefer not
to receive an incentive for reasons of safety. As such, par-
ticipants will have the option to choose to receive their
incentive amounts as an Amazon gift card emailed to
them, as a donation to a charity, or neither. ey will not
be otherwise compensated for their participation. Par-
ticipants will not incur additional costs to take part in the
study beyond what they already pay for Internet and text
messaging.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 10 of 21
Ybarraetal. Trials (2025) 26:9
Survey reminders
To increase initial enrollment rates, participants will
receive automated and then personalized outreach to
encourage the sending of one’s HIV test results, and to
complete survey assessments.
Inviting all youth tocomplete thenextsurvey follow‑up
irrespective ofprevious non‑response
To maximize data, participants will be contacted at each
data collection period irrespective of their participation
in prior follow-ups, unless they have withdrawn from the
study. Participants who do not complete the online sur-
veys within 1week may be given the opportunity to com-
plete a brief text message survey for a smaller incentive
amount. is methodology has been successfully imple-
mented in our previous studies. Indeed, this brief survey
can sometimes serve as a gateway to the full-length sur-
vey for non-responders [55].
Constantly updating contact information
We anticipate we will be able to stay in contact with youth
throughout the 15-month study period (i.e., 5-month
intervention + 12-month follow-up). In our previous stud-
ies, we have found social media to be particularly useful in
re-connecting with participants and updating their contact
information and will similarly plan to use social media to
stay in contact with participants in the current study.
Fidelity monitoring
Text messaging is associated with high fidelity because
everyone receives the content in the same order; whereas
in-person programs are subject to the facilitator’s choices
about what and when content is discussed. It is possi-
ble that unexpected technology problems may affect the
sending of the messages. To quickly identify and address
such problems, we will monitor project messages daily.
Retention strategies
To increase initial enrollment rates, participants will
receive automated and then personalized outreach to
encourage the sending of one’s HIV test results. We antic-
ipate a 70% test confirmation rate at intervention end and
12-month follow-up. We anticipate an 80% response rate
to each online survey, consistent with prior work [29].
Data management
Missing data
e primary anticipated reason for missing data is attri-
tion due to loss to follow-up, however, also may occur
within a case (skipping certain questions). is includes
those who complete a self-reported follow-up survey
but do not upload a photo confirming the result of their
OraQuick home-based test at either intervention end or
12-month post-intervention follow-up. Our statistical
methods will employ full-information maximum likeli-
hood or incorporate the expectations-maximization algo-
rithm so that all randomized individuals can be included
in an analysis even if they are lost to follow-up and/or
have partial missing data. ese methods are robust to
data that are missing at random [52, 56].
We will conduct sensitivity analyses to contextualize
how assumptions about the missing data mechanism
influence our understanding of intervention impact. To
do so, we will follow recommendations made by Leacy
etal. [57] and also analyze data with missing data coded
as failure and with missing data coded as success.
Condentiality
Safe andprivate data collection
We believe that delivering content directly to young peo-
ple’ cell phones creates a more private “space” than other
types of programming (e.g., those which are delivered
in-person).
To protect participants’ privacy, RCT data will be col-
lected via online surveys. is reduces the number of
people who view the data and increases self-disclosure on
sensitive topics. We will password protect access to the
data. We will ask youth to upload a photo oftheir HIV /
STI test to the secureonline survey platform and encour-
age them to delete the photo from their phone afterwards
(see “Secure Electronic Transmissions and Storage of
Data” below for more information).
Participants will access surveys with a personalized
link. Dr. Ybarra will oversee data collection, with the pro-
ject coordinator coordinating with the software develop-
ers and biostatisticianto continuously monitor the data.
At any time during the online surveys that participants
are asked to complete, they will have the option to pause
or stop the survey and return to it at a later time (i.e., if
they choose to provide an email address, we will send
them a link to re-enter their specific survey). We will
emphasize to the respondent that this option can be uti-
lized if he feels that the space where he is completing the
survey is no longer private.
At the beginning of each survey,we also will ask sur-
vey respondents: (1) Are you in a space that feels private?
and (2) Are you in a place where you feel comfortable
answering questions honestly? ose who say “no” to
either question will be advised that taking the survey
somewhere private and safe is important. ey will have
instructions on how to pause the survey and resume it
later if they would prefer.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 11 of 21
Ybarraetal. Trials (2025) 26:9
Data privacy
We will assign each participant a random unique identi-
fier in the dataset, stripped of all personal information
to protect confidentiality. Datasets used for analysis will
contain project identification numbers, but not names or
any other identifying information such as phone num-
ber or email address. We store identification information
separately from the responses provided by participants.
Collaborators will receive data stripped of personal
identifiers. To ensure complete confidentiality, we will
restrict access of the key linking personal identifiers to
usernames and passwords to main program staff. Dr.
Ybarra will oversee the data storage and reporting pro-
cedures. Reports will not identify individual participants;
they will only use aggregated data.
It is possible that participants may lose their confiden-
tiality if someone intercepts their phone. Study staff will
offer instructions to participants on how to password
protect their phones to limit access by others. Staff will
also encourage participants to disable text message noti-
fications that may appear on a phone screen and to delete
any messages from their phone that they do not want
anyone else to see.
Participants may lose their confidentiality if someone
intercepts the shippingbox that has their HIV test, or
the HIV test itself. We will encourage youth tohave the
packagemailed to a safe place, including the address of
a friend or family member if necessary. ey also will
be given the tracking information so they can divert the
package if need be. We also will remind people to safely
and securely dispose of the test once they have uploaded
a picture to the online survey platform (e.g., the survey
thank you message will encourage them to do so).
No additional contact information (e.g., additional
phone number, friend’s number, physical address)
beyond their cell phone number and email addresswill
be required to enroll. Participants will only provide addi-
tional contactinformation if they choose to and can do
so safely. Our team has extensive experience using this
contact information in a manner that is sensitive to the
privacy needs of youth participants.
Secure electronic transmissions andstorage ofdata
Data are located on dedicated servers at both Digital
Ocean and Liquid Web data centers. Both data centers
provide strict security compliance ensuring both physi-
cal and network security. Both servers are continuously
monitored to ensure 100% uptime.
e dedicated server facility security includes:
• 24/7/365 Manned Facilities
• CCTV Security Cameras Covering Inside, Outside
and All Entrances of Data Centers
• Site Entrances Controlled By Electronic Perimeter
Access Card System
• Sites Remotely Monitored By 3rd Party Security
Company
• Entrances Secured by Mantraps with Interlocking
Doors
• SSAE-16 & HIPAA Compliant, Safe Harbor Certified
e data centers are equipped with redundant tier
1 bandwidth, ensuring minimal latency and fast con-
nections to all points of the global Internet. Datacenter
access is strictly limited to technical staff. Electronic
security systems control datacenter access and are
accompanied by a full complement of motion-detecting
security cameras that monitor the entire facility.
All data are password protected with strong encrypted
passwords and is transmitted securely using SSL (TLS)
128-bit encryption across the Internet (HTTP). SSL pro-
vides front-end users with the assurance of access to a
valid, “non-spoofed” site and prevents data interception
or tampering with sensitive information. e 128-bit
encryption is the preferred security level of government
and financial institutions. To ensure against the remote
possibility that an intruder gains access to stored data, all
data stored are protected with strong passwords that are
also encrypted, making use of any acquired data nearly
impossible. Any Personal Identifying Information (PII)
is securely encrypted and stored separately from study
data. All access to participant data is limited to access via
a secure VPN network, making it impossible to access
otherwise.
Our data are backed up daily to an external hard drive.
We also have extensive server-hardening, firewall protec-
tion, brute force detection and evasion, denial-of-service
attack prevention/protection, and conduct daily security
audits and monthly vulnerability scans.
Plans forcollection, laboratory evaluation, andstorage
ofbiologic specimens forgenetic ormolecular analysis
inthis trial/future use
e HIV test specimens are self-collected by participants
who are using the OraQuick diagnostic test. e test is
self-administered and results are available to users at the
point of specimen collection within 20min of complet-
ing the test. Participants will be asked to self-report their
resultsby uploading a photo of the test. Only those that
are unreadable will be confirmed via laboratory result
at a local health clinic unaffiliated with the study. Each
individual test taker will dispose of their test and testing
components per OraQuick instructions. us, we have
no plans to conduct laboratory evaluation, and/or store
specimens for any current or future use.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 12 of 21
Ybarraetal. Trials (2025) 26:9
Statistical methods
Statistical methods forprimary andsecondary outcomes
Our primary analyses will be based on the intention to
treat (ITT) principle, with all randomized participants
included in the analyses regardless of the amount of data
they contribute. Analyses will be based on the initial
treatment assignment and not on the treatment eventu-
ally received.
Main outcome: OraQuick‑conrmed HIV incidence
We base our choice of when to measure HIV incidence
using OraQuick home-based tests so that we are able to
measure proximal (i.e., intervention end) and distal (i.e.,
12months post-intervention) impacts of the intervention
on incidence. Distal impact will be our main analysis of
interest.
To do so, we will utilize Kaplan–Meier survival anal-
ysis and the log-rank test of differences given experi-
mental arm assignment through the 12-month post
intervention end. In survival analysis, we are interested
in estimating two systematically related probabilities:
the hazard probability (HP) and the survival probabil-
ity (SP). In a discrete-time framework, HP is the ratio of
individuals who report, for example, a positive HIV test
at a particular time point divided by the number of indi-
viduals who were at-risk and did not report a positive
HIV test at the prior time point. SP refers to the prob-
ability of an individual surviving at least until a given
period without reporting a positive HIV test, given that
he has survived the earlier time period. Both probabili-
ties can be plotted to provide information about periods
of greatest risk over time. e DTS model is particu-
larly amenable to incidence analyses as it can appropri-
ately account for censoring individuals after they drop
out of the study prior to 12-month follow-up (i.e., right
censoring).
If treatment (treatment = 1, control = 0), were esti-
mated to have a hazard probability of 0.5, we would say
that the likelihood of an incident HIV test at interven-
tion end for the intervention group was half the proba-
bility of a positive HIV test among the control group. In
secondary analyses, we will us cox-proportional hazard
models to explore the potential impact of the follow-
ing baseline variables: age, race, ethnicity, region of the
country, sexual identity, history of HIV and STI testing
and PrEP use.
We will examine potential mediators of the treatment
effect on the primary study outcomes such as motiva-
tion to use PrEP. We will test mediation using a prod-
uct of coefficients approach with bootstrapped standard
errors (10,000 bootstrapped samples). is will allow us
to estimate path coefficients: a path (effect of treatment
assigned on changes in PrEP motivation), b path (effect
of PrEP motivation on HIV incidence) and their prod-
uct a*b (indirect effect of treatment on HIV incidence
through PrEP motivation).
To understand how program components may impact
the targeted outcomes, we also will examine whether
program appraisal, program dosage, and process meas-
ures are associated with the intervention impact on HIV
incidence among intervention participants.
As a planned secondary analysis, we will estimate the
log-odds of the intervention impact on confirmed HIV
incidence at intervention end. is will inform whether
difference in incidence was detectable more proximally.
Main outcome: self‑reported HIV incidence
In addition to test-confirmed HIV incidence, we will col-
lect self-reported HIV incidence at all time points. is
will maximize the amount of data we are collecting on
this measure and will give us an opportunity to examine
the impact that self- versus test-confirmed results impact
the conclusions we draw about intervention impact. e
self-report outcome will include data from all post-rand-
omization assessments.
Figure 2 shows a comprehensive discrete survival
model which includes regression paths for direct effects
(X, Y), moderation effects (X*Y), and mediating effects
(M). We will use generalized linear models with a logit
link to examine the effect that covariates have on the
timing of the positive self-report parameterized by its
effect on the log hazard odds of an event during a given
time interval. us, a covariate’s effect on the likelihood
of event occurrence is described in terms of the hazard
odds ratio (hOR).
Constraining covariate effects to be time-invariant will
make the effects on the hazard probability identical for
each time interval. In other words, the hOR is constant
over time (i.e., proportional hazard odds assumption).
When this constraint is relaxed, the covariate effects are
permitted to be time-variant. For example, a time-vari-
ant effect of treatment status indicates that the odds of a
positive self-report in the intervention compared to the
control group changes over time. As with test-confirmed
HIV incidence, we will test effect moderation for self-
reported incidence using interaction terms (e.g., race X
experimental arm) and mediation using the product of
coefficients approach, as above.
Assessing therelative benet ofbiological outcomes
versusself‑report
ere is an assumption that biological outcomes are
needed to identify the “true” impact of an intervention
because of inaccurate self-reporting. Based upon research
in other areas [58, 59], there seems to be little reason to
believe that there would be differential reporting by arm
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 13 of 21
Ybarraetal. Trials (2025) 26:9
(i.e., that the intervention arm would be less likely than
the control arm to report an observed positive test)—
especially in interventions such as Project SHAG where
participants are blinded to their study arm. If true, we
would expect the relative magnitude of intervention
impact to be the same for test- and self-reported HIV sta-
tus. Perhaps, however, people tend to over-report their
test results when self-reporting and the test-confirmed
report is a more sensitive measure. If true, then we would
expect the actual intervention impact to differ for the
test- and self-reported HIV status. We will explore inter-
rater reliability between self-report and OraQuick results
across arms.
In addition, we will calculate levels of interrater agree-
ment (kappa) to quantify agreement between self-report
and lab report at intervention and study end. We will use
multinomiallogistic regression modeling to explore asso-
ciations between participant characteristics and discord-
ance between self-reported and OraQuick confirmed
HIV incidence with 3 outcomes: (1) OraQuick-confirmed
and not self-reported, (2) self-reported and no OraQuick
report, (3) OraQuick-confirmed and self-reported.
Main outcome: number ofself‑reported STIs
Because participants can have a positive STI result more
than once either for the same or a different type of STI
[60, 61], we will use a count regression approach to model
the impact of the intervention on STI infections over
time [62]. Using generalized linear mixed models with
a log link function and Poisson distribution, the model
will estimate the incident rate ratio for STI infections for
those in the intervention versus control group. A random
effect for individual will account for the repeated meas-
ures throughout the study. If assumptions of the Poisson
model are violated (e.g., overdispersion), a negative bino-
mial model and if necessary zero-inflated models will be
considered as alternatives. We will also conduct media-
tion and moderation analyses as described above.
Main outcome: pre‑exposure prophylaxis (PrEP) /
post‑exposure prophylaxis (PEP) uptake
We will evaluate intervention impact on PEP/PrEP
uptake using generalized linear mixed models using a
using the binomial distribution and logit link function.
ese models will include random effects to account for
repeated measures within individuals, as above. Utilizing
the 7 data collection timelines, including baseline, mod-
els will (1) characterize the temporal trend of PrEP/PEP
use between baseline and 12-month follow-up, and (2)
assess whether these trends vary by intervention status.
Mediation and moderation will be explored as described
above.
Interim analyses
Monitoring ofText Buddy communication
Text Buddies are intervention participants whom we will
pair together and encourage to talk to each other about
program content throughout the intervention. We posit
that practicing new behaviors with a Text Buddy will
reinforce HIV preventive behavior change over time.
Fig. 2 Planned SHAG analysis: discrete survival model
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 14 of 21
Ybarraetal. Trials (2025) 26:9
However, balancing this with the privacy needs of our
participants, who likely will not want a person unknown
to them to have their cell phone number, is important.
To achieve this, all Text Buddy messages will be routed
through the study server, replacing the need to exchange
cell phone numbers. Buddy conversations will be saved
in a password-protected file. For analysis, we will keep a
count record to reflect the number of messages sent by
each participant.
We will have a safety plan for Text Buddies. It is pos-
sible that interactions between Text Buddies will be
unhealthy (e.g., encouragement of risky sexual behav-
ior). We will monitor the interactions between Buddies
daily during the RCT to determine if this occurs. We also
will block messages that contain key words (e.g., contact
information) for review before they pass through to the
Buddy. Any concerning message content will be elevated to
the PI within 24h. We will suspend the participant’s Text
Buddy access immediately, and study staff will contact
them by telephone to resolve the issue.
Methods foradditional analyses
Secondary outcomes: (1) information about and (2) moti-
vation for uptake of PrEP; (3) STI Testing; and (4) the
impact of the intervention on mental health indicators.
We will use generalized linear mixed models (with an
identity link function) to test whether the intervention
is associated with higher scores of PrEP information and
PrEP motivation compared to the control. ese models
will utilize all seven time points, with time as an inter-
action effect to determine whether the score differences
attenuate over time. Specific contrasts will be made to
test for intervention effects at the end of the intervention
and at 12months post follow-up.
Given a focus of the intervention on reducing stigma,
using substances during sexual episodes, and increas-
ing social support to affect HIV preventive behavior and
positive outcomes, we will examine whether those in the
intervention have greater improvement on these mental
health indicators over time than do those in the control.
ese models will be estimated in the same manner as
PrEP information and motivation.
Unique statistical analysis challenges posed byanHIV
incidence endpoint andapproaches tomanage other
expected study outcomes
Because HIV incidence is a relatively rare event, it is
challenging to have sufficient power to detect differ-
ences by experimental arms for this endpoint. is chal-
lenge is compounded by the fact that sero-positivity can
sometimes take up to 3months post-infection to detect.
Nonetheless, even when we account for the possibility
that we will miss youth whose HIV is not yet detectable,
our observation period (from baseline to 12-month post-
intervention follow-up) is 14 months. However, we
submit that the study remains adequately powered to
measure beyond 1-year incidence.
An incidence endpoint also requires that participants
are sero-negative at baseline. is makes it challenging to
recruit youth: A large number of youth who would oth-
erwise take part if an HIV test result verification were
not required will decline to assent/consent. We also will
experience attrition between the consent/assent and ran-
domization period because participants who are mailed
the test do not take it, or they take it but choose not to
upload the result in the baseline survey platform. To
manage this, our plan assumes 25% attrition from assent/
consent to randomization; we plan to assent/consent
6777 youth to randomize a sufficiently large sample, i.e.,
n = 5000.
Another challenge is how to treat youth who self-
report HIV positivity but do not provide confirmation of
the test. ose who provide photo-verification of a posi-
tive result are censored from subsequent data collection
efforts. We will conduct sensitivity analyses to under-
stand the implications of including these self-reported
positive results as positive OraQuick-verified analyses
versus treating them as negative.
Because of these noted challenges, it is even more
important to manage other expected study outcomes
in a way that increases the likelihood that true differ-
ences in experimental arms (i.e., intervention impact)
are detected. To this end, we also will be analyzing self-
reported HIV incidence; this will help address potential
under-reporting by youth who do not want to upload
their test result, especially a positive one. We also will
analyze study endpoints that are expected to have higher
prevalence and also be less stigmatizing to report the
number of STIs, and uptake of PrEP and PEP. As proxi-
mal indicators of risk for HIV acquisition, these outcomes
will further help contextualize the effect the intervention
may have on HIV incidence.
Methods inanalysis tohandle protocol non‑adherence
andany statistical methods tohandle missing data
Given that the program will be delivered via technology,
protocol non-adherence could occur if there are tech-
nological challenges (e.g., systemic non-sending of text
messages) or if participants purposefully misreport their
PII to appear eligible when they are not. We will closely
monitor software performance and quickly problem
solve any technology issues to ensure that technology-
related disruptions of message delivery do not affect the
data. Any challenges will be noted by date and time in
a “field blog” so that if systemic issues are noted, people
affected by these issues can be examined separately in the
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 15 of 21
Ybarraetal. Trials (2025) 26:9
analyses to determine if their outcomes are different from
unaffected youth. To reduce the likelihood of enrolling
people who are ineligible, we will use logic in the screener
to redirect ineligible people to a “thank you” page. We
also will identify people with duplicate email addresses,
phone numbers, and/or mailing addresses. Given the
highly interactive nature of the program, we also will note
interactions with participants that seem age incongruent
(e.g., language used) and reach out to anyone flagged by
research staff to further confirm their identity.
We will conduct sensitivity analyses to contextualize
how assumptions about the missing data mechanism
influence our understanding of intervention impact. To
do so, we will follow recommendations made by Leacy
etal. [57] and also analyze data with missing data coded
as failure and with missing data coded as success.
Plans togive access tothefull protocol, participant‑level
data, andstatistical code
We provide full public access to the protocol through
this document and upon request to clarify or offer more
detail on specific study procedures. We will provide
de-identified participant level data upon request when
provided with a clear rationale and detailed analytic
plan, e.g., for use in a meta-analysis. While our protocol
includes details on analysis, we do not plan on releasing
statistical code.
Oversight andmonitoring
Composition ofthecoordinating center andtrial steering
committee
e research team will comprise the coordinating center.
e team includes the PI, a project coordinator, research
assistants, as well as multiple software developers and
other technology support personnel. e team will
meet weekly during the trial implementation to discuss
recruitment, enrollment, intervention implementation,
and documentation as well as any protocol deviations or
adverse events should they arise. e trial steering com-
mittee comprises the PI, project coordinator, and 3 pro-
ject co-investigators. is group will meet periodically to
discuss project implementation.
Composition ofthedata monitoring committee, its role
andreporting structure
e PI will be responsible for all aspects of the project
with respect to intervention design, data collection, and
use of data. Pearl IRB will approve all aspects of the study
prior to commencing data collection. Informed youth
assent or adult consentwill be obtained from all partici-
pants. Participants will have access to referral informa-
tionto resources that provide support 24 h a day (e.g.,
Trevor Project) as well as phone numbers for the study
administrators (i.e., PI and IRB).
Data monitoring
Study staff will monitor the quality of the evaluation
as it occurs, using our performance measure data. For
example, we will monitor our adherence to the proposed
timeline by tracking our enrollment rates, the percent-
age of youth who assent but do not complete the base-
line survey (i.e., the baseline response rate), and our
follow-up response rates. If response rates are lower
than anticipated, the project team, including consultants,
will convene and brainstorm ideas to invigorate survey
completion. Intervention participation will be measured
through the intervention’s weekly level-up questions,
the amount of interaction with their Text Buddy, the
acquisition of badges, etc. Survey data will be monitored
continuously to quickly identify any problems (e.g., pro-
gramming of the skip patterns, unexpectedly high “do not
want to answer” rates for questions).
We have developed, and will refine, an online moni-
toring interface for this project that allows project staff
to monitor the program messages sent to participants,
the messages that participants send to the program,
and participants’ progression through the program. Any
problems with program functioning will be immediately
elevated to the technology team to resolve.
Adverse event reporting andharms
We will promptly report unanticipated problems to the
IRB and appropriate institutional officials of (i) any unan-
ticipated problems involving risks to participants or oth-
ers, or any serious or continuing noncompliance with this
policy or the requirements or determinations of the IRB;
and (ii) any suspension or termination of IRB approval.
e purpose of prompt reporting is to ensure that appro-
priate steps are taken in a timely manner to protect other
participants from avoidable harm. To be specific, we will
promptly report to the IRB:
• Deviations and violations in accordance with local,
institutional, or protocol-specific guidelines/proce-
dures
• Changes increasing the risk to subjects and/or sig-
nificantly affecting the conduct of the trial
• Adverse events (definition per International Council
for Harmonisation, Good Clinical Practice, and the
Food and Drug Administration) as specified in the
protocol or by IRB policy
• New information that may adversely affect the safety
of the participants or the conduct of the study
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 16 of 21
Ybarraetal. Trials (2025) 26:9
In general, we will define “prompt” as, and report
accordingly when possible:
• Unanticipated problems that are serious adverse
events will be reported to the IRB within 1week of
the investigator becoming aware of the event
• Any other unanticipated problem should be reported
to the IRB within 2weeks of the investigator becom-
ing aware of the problem
• All unanticipated problems should be reported to
appropriate institutional officials (as required by an
institution’s written reporting procedures), the sup-
porting agency head (or designee), and Office for
Human Research Protections within 1month of the
IRB’s receipt of the report of the problem from the
investigator
e PI will review the project progress and the col-
lected data to ensure that potential adverse effects, if they
occur, are identified and reported to the IRB. Any action
recommended by the IRB will be implemented immedi-
ately in order to minimize further risk. All notifications
will be done via email.
Frequency andplans forauditing trial conduct
We have a data safety and monitoring plan as described
above re: interim analyses. Our team will review progress
on achieving enrollment goals each week and will moni-
tor intervention delivery to document and address any
technology errors or failures on a daily basis during study
implementation.
Plans forcommunicating important protocol amendments
torelevant parties (e.g., trial participants, ethical
committees)
We do not anticipate any changes to eligibility criteria,
outcomes or analytic plans that are described here. How-
ever, if any of these do occur, we will submit amendments
to our IRB for review and approval and will make modifi-
cations to our ClinicalTrials.gov registration, if necessary.
Such amendments are not communicated to participants
unless it would affect their own eligibility and continued
enrollment in the trial.
Dissemination plans
We have a multi-tiered dissemination plan. For research-
ers, we will publish our findings widely in peer-reviewed
journals. We also will present findings at at least two pro-
fessional conferences (e.g., AIDS). We will not utilize pro-
fessional writers, and authorship eligibility guidelines will
conform to guidelines published by Fontanarosa etal., in
2017 [63]. For both researchers and public consumers,
we will create a webpage on the CiPHR website where
people will be able to download all study materials, jour-
nal, articles, and media mentions about the study free of
charge. Finally, we will register with ClinicalTrials.gov.
Discussion
Although this proposal is highly significant and innova-
tive, it is not without limitations: (1) While 12months
is a sufficient follow-up period, 24 months would have
been preferred. We chose to preference sufficient time
for recruitment and enrollment to ensure a sufficient
sample size over a longer follow-up period with a smaller
sample. We also deemed it unlikely that differences in
incidence, which is a low occurring event, would be
detectable through 24 months without ongoing inter-
vention into the second year. (2) It bears noting that the
OraQuick home-based HIV test may not detect infection
that has occurred within the past 3months. If we “adjust”
our surveillance period for this possibility, we will still be
measuring prevalence across 14 months from baseline
to 12-month follow-up. (3) It also would be preferred
to confirm HIV sero-status using blood assays. Based
upon the conversations we had with colleagues, how-
ever, we believe the impact this type of test would have
on our response rate would make it impossible to reach
our recruitment and retention goals. (4) Additionally,
it would potentiallybe transformative if we could mail
PrEP to our participants. at said, even if we could iden-
tify a way to remotely prescribe to participants—includ-
ing those under 18years of age, the sustainability of such
a program after the study ends is highly questionable,
and the benefit of technology-based interventions is their
low cost and scalability. It also may bear noting that an
intervention centered on the mass distribution of PrEP is
less about behavioral interventions that impact HIV inci-
dence and more about whether people will uptake PrEP
if it is proactively mailed to them. e data are very clear
on PrEP’s preventive impact. Understanding the impact
that increased access would have on incidence is an
important research question, but one the current study.
(5) Some also may have concerns about the feasibility and
acceptability of such an intense, long intervention. Our
previous work suggests reason for optimism. (6) It also
bears noting that not all youth will have unlimited Inter-
net bandwidth, reducing their access to videos. e inter-
vention text messaging content is written to stand alone
however, so we believe that this will not be a significant
limitation. (7) Finally, gender diverse youth are excluded.
is is because the time necessary to develop content
that is appropriately gender affirming is more than what
is afforded in the first year of the grant.
e rigor of the proposed study is high: Random
assignment eliminates the possibility of youth being
purposefully assigned to a particular study arm. e
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 17 of 21
Ybarraetal. Trials (2025) 26:9
attention-matched control will reduce the likelihood that
behavior change, if detected, is due to the “attention”
youth received by the daily messages. Moreover, the pop-
ulation-based focus on the intervention design and pilot
testing increases the generalizability of findings beyond
cisgendersexual minority boys and young men living in
one city or a particular region, and increases the likeli-
hood that the program could feasibly be disseminated at
the public health level. e manner in which youth are
recruited will increase the likelihood that the sample
reflects youth who might use the intervention if it were
available publicly (e.g., the lack of incentives for the base-
line survey, the lack of mention of incentives in recruit-
ment ads), while also increasing the likelihood that they
are who they say they are (e.g., telephone enrollment).
In conclusion, because of our population-based
approach to finalizing and testing the intervention, if
findings are positive, the intervention can be quickly
made publicly available to affect HIV incidence at the
population level.
Trial status
Protocol date: 5/31/2023. Date recruitment began: Janu-
ary 15, 2024. Approximate date recruitment will be com-
pleted: January 15, 2027.
Appendix
PROJECT SHAG RCT CONSENT FORM
STUDY TITLE: HARNESSING THE POWER OF
TEXT MESSAGING TO REDUCE HIV INCIDENCE
IN ADOLESCENT MALES ACROSS THE UNITED
STATES
FUNDER: NATIONAL INSTITUTE OF CHILD
HEALTH AND HUMAN DEVELOPMENT GRANT
NUMBER: U01HD108738
SPONSER: CENTER FOR INNOVATIVE PUBLIC
HEALTH RESEARCH (CIPHR) PRIMARY INVESTIGA-
TOR: MICHELE YBARRA, MPH PHD
e Center for Innovative Public Health Research has
developed two sexual health programs for guys who are
into guys. You are being asked to take part in the rand-
omized controlled trial of the two programs to see which
one works better. e programs will be sent through text
messaging. Program text messages talk about things like
safely having sex with guys, ways to prevent HIV and
other STIs, and using condoms and PrEP.
is research study is sponsored by the National Insti-
tutes of Child Health and Human Services.
Procedures
Your participation will last about 18months.
You are one of about 5,000 guys 13–22years old being
invited to take part in the Project SHAG study. SHAG
stands for: Sexual Health Advocacy for Guys.
ere are two different text messaging programs that
we are testing. We do not know which program is bet-
ter at promoting healthy sexual behavior. You will be ran-
domly assigned to either program. is means you have
an equal chance of being in either program. We will not
tell you which program you are assigned to until after
everyone has finished the program.
If you choose to take part in the research study, here’s
what we will ask you to do:
1. Complete an online survey and an HIV test before
you start the program. We will mail you the test. You
can do it anywhere and anytime that is safe for you.
We ask you to upload a picture of the results of the
test to our secure study server so that we are on the
same page about your result. Only the research team
will have access to the picture; we take your privacy
seriously.
2. Once you finish the survey and upload your test
results, you will be officially enrolled. You will receive
between 5–10 text messages every day for 9weeks.
You may also be randomly matched to a “text buddy”,
another guy in this study, who you will be able to talk
to about the things that you are learning in the pro-
gram.
3. After 9weeks, the daily text messages will stop, and
we will ask you to complete another online survey.
4. We will then send you a couple of texts per week for
the next 3months.
5. After that, you will receive a “review week” where you
will receive 5–10 messages again each day. After the
review week, we will send you another HIV test and
a survey link and ask you to upload your test results
and do the online survey.
6. Over the next 12 months, we will ask you to do
4 more online surveys, once every 3 months. At
12months, as part of your last survey, we will send
you a final HIV test and ask you to upload a picture
of the results to the secure Project SHAG server.
e only cost for you to take part in this study will be
costs that you already pay for text messaging and to go
online. We will pay for the HIV tests and the shipping
costs.
Incentives
You can receive up to $270 for taking part in this study.
Here’s how it breaks down:
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 18 of 21
Ybarraetal. Trials (2025) 26:9
For doing the online surveys, you’ll get:
• $15 for completing the ‘core’ intervention end survey
at 9weeks
• $15 for completing the intervention end survey at
5.5months
• $25 for completing the 3-month survey after the end
of the intervention
• $25 for completing the 6-month survey after the end
of the intervention
• $25 for completing the 9-month survey after the end
of the intervention
• $30 for completing the 12-month survey after the
end of the intervention
You can also earn a bonus if you upload your HIV test:
• You can earn a $30 bonus for uploading your fist test
result in the first survey
• You can earn a $45 bonus for uploading your second
test result at the end of the intervention, 5.5months
later
• You can earn a $60 bonus for uploading your third
test result at the end of the study
You may choose not to upload a photo of your results.
In this case, you will not get the bonus. It is also possible
a different bonus may be offered during your time in the
study.
Your incentives and bonuses will be sent to you as an
Amazon gift card to the email address you give us. You
will also have the option to donate your incentive to a
charity, or choose not to get an incentive at all.You may
be asked to confirm your identity by verifying your per-
sonal information at any time during the study. You can
do so by joining a video call with study staff or sending
us a copy of your government issued ID, among other
methods. Not confirming your identity may result in
being removed from the study and not getting paid your
incentives.
Risks and Discomforts
It is possible that you will learn that you have HIV. is
could be very upsetting. Some guys experience problems
with family or have emotional difficulty learning that they
are HIV positive. If you are worried about whether you
can stay safe if you learn you have HIV, this might not be
the right time for you to be in this study.
It is also possible that your privacy will be broken. For
example, someone might see the shipping package, the
HIV test, the program text messages or the online survey
on your device and ask you about it. We want to protect
your privacy as much as possible so it is very important
that you have the HIV test mailed somewhere that is safe
for you, and that you take the test in a private place.
It also is important that you receive the program text
messages and take the surveys on your own private
device – not one you share with others. If your phone
is linked to another device like a family-shared tablet,
maybe think about being sure that the messages don’t
scroll on this other device.
Survey questions we ask might make you feel uncom-
fortable. If this happens, you can select ‘Do not want to
answer;’ leave the survey and not answer the question;
or stop being in the study completely. Please know that
some questions, such as your birthday, are required if you
want to take part.
It also is possible that something else might happen
that we have not thought about yet.
Benefits
You may benefit by knowing your HIV status. If you
are positive, you can start lifesaving treatments. If you
are negative, you can continue making choices to reduce
your risk of getting HIV, like using condoms or getting on
PrEP (a pill or shot that reduces your chance of getting
HIV). You may also learn ways to have a healthy sex life.
Rights of Refusal and Withdrawal.
You can choose to be in the study or not. If you decide
not to be in the study, that is OK; nothing bad will
happen.
You can choose to stop being in the study at any time,
even if you have already started. If you decide you do
not want to be in the study after it has started, just let us
know by texting us at 714- 203–2755.
Your time in the study may also stop at any time for
any reason, such as, the sponsor or the study investigator
decides to stop the study.
Confidentiality
We will keep a copy of your answers after the study
ends so that we can look at them later. We will only share
research data where we have removed anything that we
think would show your identity. ere still may be a small
chance that someone could figure out that the informa-
tion is about you. Examples of sharing include:
• Publishing results in a book or journal.
• Adding results to a federal government database.
• Using research data in future studies, done by us or
by other scientists.
• Representatives at the Department of Health and
Human Services and Pearl IRB also may request
access to the study data.
We will analyze your responses to the surveys and the
messages to determine which program is better. If you
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 19 of 21
Ybarraetal. Trials (2025) 26:9
are matched with a Text Buddy, we may also analyze your
conversation to better understand the lives of guys today
and also identify ways to improve Project SHAG. We may
also use study data to look at another research question
that we have not thought of yet.
Aside from the sharing of research data we note above,
we will not tell anyone what your HIV test result is, or
what your answers are on the surveys.
To help us protect your privacy, we have obtained a
Certificate of Confidentiality from the National Institutes
of Health. is Certificate means that we can keep your
information private even if we get a court order telling us
to share your information. We will use this Certificate to
fight demands for your information unless you tell us you
want us to share the information. In the unlikely event
that you tell us that you are being harmed or harming
others, then under applicable law, we may be required to
report this information to the appropriate authorities.
Questions and Contact Numbers
e researchers do not have a conflict of interest in this
study.
If you have questions about the study or any concerns
about the study questions, please contact:
• Dr. Michele Ybarra toll-free at 1–877-302–6858 ext.
801 or Michele@InnovativePublicHealth.org.
• If you have questions about your rights as a partici-
pant in this study, or if you feel that you have been
harmed in any way by taking part in this study, please
contact Pearl IRB:
o By mail: Study Subject Adviser Pearl IRB 29 East
McCarty Street, Suite 100 Indianapolis, IN 46225
or
o Call: 317–899-9341 or
o By email: info@pearlirb.com
An IRB is a group of people who review research
studies to protect the rights and safety of research par-
ticipants. Please reference the following study title when
contacting the Study Subject Adviser: Project SHAG.
Here are some resources that you may find helpful:
• If you would like to find a clinic where you can get
tested for HIV, go here: https:// gette sted. cdc. gov/
• Here is information about HIV: https:// www. cdc.
gov/ hiv/ basics/ whati shiv. html, and resources for
people who are living with HIV: https:// www. cdc.
gov/ hiv/ basics/ livin gwith hiv/ resou rces. html
• You can always talk to someone at the Trevor-
Project for support. ey have a 24-h hotline for
LGBT + youth: 1–866-488–7386, or text ‘START’ to
678678.
• If, at any time, you think about hurting yourself,
please contact the National Suicide Prevention Hot-
line at: 988. ey can help.
Do you want to take part in this 2-year study? No.
Yes.
[If the person agrees to take part in the study:]
Can we contact you after the study ends if we have
questions for you about the study? No.
Yes.
[Asked of everyone].
Would you like us to tell you about other studies that
you might be eligible for? No.
Yes.
[If the person declines to take part in the study:]
We respect your decision. So that we can better design
studies in the future, could you please share why you
would not like to be part of Project SHAG?
[non-mandatory text box].
Abbreviations
CiPHR Center for Innovative Public Health Research
FB Facebook
FDA Food and Drug Administration
G2G Guy2Guy
HIV Human immunodeficiency virus
IG Instagram
IRB Institutional Review Board
MTL Molecular Testing Labs
PEP Post-exposure prophylaxis
PI Principal Investigator
PII Personal Identifying Information
PrEP Pre-exposure prophylaxis
RCT Randomized controlled trial
SGM Sexual and gender minority
SHAG Sexual Health Advocacy for Guys
SPIRIT Standard Protocol Items: Recommendations for Intervention Trials
STI Sexually transmitted infections
Supplementary Information
The online version contains supplementary material available at https:// doi.
org/ 10. 1186/ s13063- 024- 08540-9.
Supplementary Material 1.
Acknowledgements
We appreciate Dr. Pan Yue’s review and comments on drafts of the manuscript
and Dr. Allie White’s assistance with manuscript formatting and review.
Authors’ contributions
MY is the Principal Investigator; she conceived the study, led the proposal
and protocol development. RG and SB contributed to study design and to
development of the proposal. DF was the lead trial methodologist. All authors
read and approved the final manuscript.
Funding
Research reported in this publication was supported by the Eunice Ken-
nedy Shriver National Institute Of Child Health & Human Development of
the National Institutes of Health under Award Number U01HD108738. The
content is solely the responsibility of the authors and does not necessarily
represent the official views of the National Institutes of Health.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 20 of 21
Ybarraetal. Trials (2025) 26:9
Data availability
We will make limited datasets available to researchers with detailed proposal,
IRB supervision and the appropriate data use agreements in place.
Declarations
Ethics approval and consent to participate
Pearl IRB reviewed and approved the study protocol. The reference number
is 21-CIPH-106. Oral informed consent for those 18 years of age and above,
and assent for those under 18, will be obtained from all participants. A waiver
of parental permission for those under 18 years of age has been approved by
the IRB.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Author details
1 Center for Innovative Public Health Research, San Clemente, CA, USA.
2 Department of Public Health Sciences, University of Miami Miller School
of Medicine, Miami, FL, USA. 3 Ann & Robert H. Lurie Children’s Hospital of Chi-
cago; Northwestern’s Feinberg School of Medicine, Chicago, USA. 4 Depart-
ment of Community and Behavioral Health, Colorado School of Public Health,
Denver, CO, USA.
Received: 8 July 2024 Accepted: 8 October 2024
References
1. Centers for Disease Control and Prevention. Effective interventions.
Published 2020. https:// www. cdc. gov/ hiv/ resea rch/ inter venti onres earch/
ebis/ index. html.
2. National Center for HIV/AIDS. HIV surveillance - Adolescent and young
adults. Published 2017. https:// www. cdc. gov/ hiv/ pdf/ libra ry/ slide sets/
cdc- hiv- surve illan ce- adole scents- young- adults- 2017. pdf.
3. Centers for Disease Control and Prevention. HIV and youth. 2019;(March
16, 2020). https:// www. cdc. gov/ hiv/ group/ age/ youth/ index. html.
4. Denning P, Dinenno E. Communities in Crisis: Is There a Generalized HIV
Epidemic in Impoverished Urban Areas of the United States? U.S. Depart-
ment of Health & Human Services; 2019. http:// www. cdc. gov/ hiv/ group/
pover ty. html.
5. Centers for Disease Control and Prevention. A web-based survey of HIV
testing and risk behaviors among gay, bisexual, and other men who have
sex with men—United States, 2012. HIV Surveillance Special Report 14.
Published 2015. http:// www. cdc. gov/ hiv/ pdf/ libra ry/ repor ts/ surve illan
ce/ cdc- hiv- hssr- whbs- msm- 2012. pdf.
6. Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophylaxis for
HIV prevention in men who have sex with men. N Engl J Med. 2010;363.
https:// doi. org/ 10. 1056/ NEJMo a1011 205.
7. Thigpen MC, Kebaabetswe PM, Paxton LA, et al. Antiretroviral preexpo-
sure prophylaxis for heterosexual HIV transmission in Botswana. N Engl J
Med. 2012;367(5):423–34. https:// doi. org/ 10. 1056/ NEJMo a1110 711.
8. Baeten JM, Donnell D, Ndase P, et al. Antiretroviral prophylaxis for HIV pre-
vention in heterosexual men and women. N Engl J Med. 2012;367(5):399–
410. https:// doi. org/ 10. 1056/ NEJMo a1108 524.
9. Choopanya K, Martin M, Suntharasamai P, et al. Antiretroviral prophylaxis
for HIV infection in injecting drug users in Bangkok, Thailand (the Bang-
kok Tenofovir Study): a randomised, double-blind, placebo-controlled
phase 3 trial. Lancet. 2013;381(9883):2083–90. https:// doi. org/ 10. 1016/
S0140- 6736(13) 61127-7.
10. Fonner VA, Dalglish SL, Kennedy CE, et al. Effectiveness and safety of oral
HIV preexposure prophylaxis for all populations. AIDS. 2016;30(12):1973–
83. https:// doi. org/ 10. 1097/ qad. 00000 00000 001145.
11. McCormack S, Dunn DT, Desai M, et al. Pre-exposure prophylaxis to
prevent the acquisition of HIV-1 infection (PROUD): effectiveness results
from the pilot phase of a pragmatic open-label randomised trial. Lancet.
2016;387(10013):53–60. https:// doi. org/ 10. 1016/ S0140- 6736(15) 00056-2.
12. Centers for Disease Control and Prevention. STDs at record high, indicat-
ing urgent need for prevention. Published online 2017. Available at:
https:// archi ve. cdc. gov/ www_ cdc_ gov/ media/ relea ses/ 2017/ p0926- std-
preve ntion. html. Accessed 1 Nov 2024.
13. Centers for Disease Control and Prevention. Adolescents and young
adults. Published 2017. https:// www. cdc. gov/ std/ life- stages- popul ations/
adole scents- young adults. htm. Accessed 1 Oct 2017.
14. Pathela P, Braunstein SL, Blank S, Schillinger JA. HIV incidence among
men with and those without sexually transmitted rectal infections:
estimates from matching against an HIV case registry. Clin Infect Dis.
2013;57(8):1203–9. https:// doi. org/ 10. 1093/ cid/ cit437.
15. Hayes R, Watson-Jones D, Celum C, van de Wijgert J, Wasserheit J. Treat-
ment of sexually transmitted infections for HIV prevention: end of the
road or new beginning? AIDS. 2010;24(Suppl 4):S15–26. https:// doi. org/
10. 1097/ 01. aids. 00003 90704. 35642. 47.
16. Centers for Disease Control and Prevention. Sexually Transmitted Disease
Surveillance 2014. 2015.
17. Centers for Disease Control and Prevention. Sexually Transmitted Disease
Surveillance 2014. Atlanta: U.S. Department of Health and Human Ser-
vices; 2015. https:// www. cdc. gov/ std/ stats/ archi ve/ surv- 2014- print. PDF.
Accessed 1 Nov 2024.
18. Centers for Disease Control and Prevention. Sexually Transmitted Disease
Surveillance 2015. Atlanta: U.S. Department of Health and Human
Services; 2016. https:// www. cdc. gov/ std/ stats/ archi ve/ STD- Surve illan ce-
2015- print. pdf. Accessed 1 Nov 2024.
19. Josie L, Centers for Disease Control and Prevention. Vital signs: HIV infec-
tion, testing, and risk behaviors among youths — United States. MMWR
Morb Mortal Wkly Rep. 2012;61(47):971–6 http:// www. cdc. gov/ mmwr/
previ ew/ mmwrh tml/ mm614 7a5. htm.
20. Laumann EO, Youm Y. Racial/ethnic group differences in the prevalence
of sexually transmitted diseases in the United States: a network explana-
tion. Sex Transm Dis. 1999;26(5):250–61. https:// doi. org/ 10. 1097/ 00007
435- 19990 5000- 00003.
21. National Research Council. The Hidden Epidemic: Confronting Sexually
Transmitted Diseases. (Eng TR, Butler WT, eds.). The National Academies
Press; 1997. Available at: https:// nap. natio nalac ademi es. org/ catal og/
5702/ the- hidden- epide mic- confr onting- sexua lly- trans mitted- disea ses-
summa ry. Accessed 1 Nov 2024.
22. Centers for Disease Control and Prevention. STDs in Racial and Ethnic
Minorities. Published 2011. http:// www. cdc. gov/ std/ stats 10/ minor ities. htm.
23. Lenhart A, Ling R, Campbell S, Purcell K. Teens and Mobile Phones. Pew
Internet & American Life Project; 2010. http:// www. pewin ternet. org/ files/ old-
media// Files/ Repor ts/ 2010/ PIP- Teens- and- Mobile- 2010- with- topli ne. pdf.
24. Whittaker R, Merry S, Dorey E, Maddison R. A development and evalua-
tion process for mHealth interventions: examples from New Zealand. J
Health Commun. 2012;17(Suppl 1):11–21. https:// doi. org/ 10. 1080/ 10810
730. 2011. 649103.
25. Preston KE, Walhart TA, O’Sullivan AL. Prompting healthy behavior via
text messaging in adolescents and young adults. Am J Lifestyle Med.
2011;5(3):247–52. https:// doi. org/ 10. 1177/ 15598 27610 392325.
26. Wei J, Hollin I, Kachnowski S. A review of the use of mobile phone text
messaging in clinical and healthy behaviour interventions. J Telemed
Telecare. 2011;17(1):41–8. https:// doi. org/ 10. 1258/ jtt. 2010. 100322.
27. Cole-Lewis H, Kershaw T. Text messaging as a tool for behavior change in
disease prevention and management. Epidemiol Rev. 2010;32(1):56–69.
https:// doi. org/ 10. 1093/ epirev/ mxq004.
28. Ybarra ML, Jiang Y, Free C, Abroms LC, Whittaker R. Participant-level meta-
analysis of mobile phone-based interventions for smoking cessation
across different countries. Prev Med (Baltim). 2016;89:90–7. https:// doi.
org/ 10. 1016/j. ypmed. 2016. 05. 002.
29. Ybarra ML, Prescott T, Phillips G, Bull S, Parsons J, Mustanski B. Pilot RCT
results of an mHealth HIV prevention program for sexual minority male
adolescents. Pediatrics. 2017;140(1):e2016299. https:// doi. org/ 10. 1542/
peds. 2016- 2999.
30. Ybarra M, Goodenow C, Rosario M, Saewyc E, Prescott T. An mHealth
Intervention for pregnancy prevention for LGB teens: an RCT. Pediatrics.
2021;147(3):e2020013607. https:// doi. org/ 10. 1542/ peds. 2020- 013607.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 21 of 21
Ybarraetal. Trials (2025) 26:9
31. Ybarra M, Rosario M, Saewyc E, Goodenow C, Dunsiger S. One-year
follow-up after a pregnancy prevention intervention for LGB+ Teens: An
RCT. Pediatrics. 2023;151(4):e2022059172. https:// doi. org/ 10. 1542/ peds.
2022- 059172.
32. Ybarra ML, Agaba E, Chen E, Nyemara N. Iterative development of In This
toGether, the first mHealth HIV prevention program for older adolescents
in Uganda. AIDS Behav Published online. 2020. https:// doi. org/ 10. 1007/
s10461- 020- 02795-4.
33. Chan A-W, Tetzlaff JM, Gøtzsche PC, Altman DG, Mann H, Berlin J, Dick-
ersin K, Hróbjartsson A, Schulz KF, Parulekar WR, Krleža-Jerić K, Laupacis
A, Moher D. SPIRIT 2013 Explanation and Elaboration: Guidance for
protocols of clinical trials. BMJ. 2013;346:e7586.
34. Moser DJ, Schultz SK, Arndt S, et al. Capacity to provide informed consent
for participation in schizophrenia and HIV research. Am J Psychiatry.
2002;159(7):1201–7. https:// doi. org/ 10. 1176/ appi. ajp. 159.7. 1201.
35. Center for Innovative Public Health Research. Guy2Guy phase three: Beta
test: Safety Appraisal Questions. Published 2014. https:// innov ative publi
cheal th. org/ wp- conte nt/ uploa ds/ 2015/ 06/ Guy2G uy- Phase- 3.- Beta- test-
Safety- Appra isal- Quest ions. pdf. Accessed 5 Aug 2020.
36. Lenhart A. Teens, Social Media, & Technology Overview 2015. Pew Inter-
net & American Life Project; 2015. http:// www. pewin ternet. org/ 2015/ 04/
09/ teens- social- media- techn ology- 2015/.
37. Lopez MH, Krogstad JM, Flores A. Key facts about young Latinos, one of
the nation’s fastest-growing populations. FactTank News in the Numbers.
Published 2018. https:// www. pewre search. org/ fact- tank/ 2018/ 09/ 13/
key- facts- about- young- latin os/.
38. Ybarra ML, Price-Feeney M, Prescott T, Goodenow C, Saewyc E, Rosario
M. Girl2Girl: How to develop a salient pregnancy prevention program
for cisgender sexual minority adolescent girls. J Adolesc. 2020;85:41–58.
https:// doi. org/ 10. 1016/j. adole scence. 2020. 09. 006.
39. Dunn LB, Jeste DV. Enhancing informed consent for research and treat-
ment. Neuropsychopharmacology. 2001;24(6):595–607. https:// doi. org/
10. 1016/ S0893- 133X(00) 00218-9.
40. The University of California at San Diego Task Force on Decisional
Capacity. Procedures for determination of decisional capacity in persons
participating in research protocols. Published 2003. https:// irb. ucsd. edu/
Decis ional Capac ityAs sessm ent. pdf.
41. Greytak EA, Kosciw JG, Diaz EM. Harsh Realities: The Experience of
Transgender Youth in Our Nation’s Schools. GLSEN; 2009. http:// glsen.
org/ sites/ defau lt/ files/ Harsh Reali ties. pdf.
42. Lenhart A. Teens, Smartphones & Texting. Pew Internet & American Life
Project; 2012. http:// www. pewin ternet. org/ Repor ts/ 2012/ Teens- and-
smart phones. aspx.
43. Ybarra ML, Prescott TL, Philips GL 2nd, Bull SS, Parsons JT, Mustanski B.
Iteratively developing an mHealth HIV prevention program for sexual
minority adolescent men. AIDS Behav. 2016;20(6):1157–72. https:// doi.
org/ 10. 1007/ s10461- 015- 1146-3.
44. Ybarra ML, Holtrop JS, Prescott TL, Rahbar MH, Strong D. Pilot RCT results
of Stop My Smoking USA: a text messaging-based smoking cessation
program for young adults. Nicotine Tob Res. 2013;15(8):1388–99. https://
doi. org/ 10. 1093/ ntr/ nts339.
45. Kreuter MW, Farrell DW, Olevitch LR, Brennan LK. Tailoring Health Mes-
sages: Customizing Communication With Computer Technology (1st ed.).
Routledge; 2000. https:// doi. org/ 10. 4324/ 97813 15045 382.
46. Alden DL, Friend J, Schapira M, Stiggelbout A. Cultural targeting and tai-
loring of shared decision making technology: a theoretical framework for
improving the effectiveness of patient decision aids in culturally diverse
groups. Soc Sci Med. 2014;105:1–8. https:// doi. org/ 10. 1016/j. socsc imed.
2014. 01. 002.
47. Schmid KL, Rivers SE, Latimer AE, Salovey P. Targeting or tailoring? Maximiz-
ing resources to create effective health communications. Mark Health Serv.
2008;28(1):32–7 http:// www. ncbi. nlm. nih. gov/ pmc/ artic les/ PMC27 28473/.
48. Rodgers A, Corbett T, Bramley D, Riddell T, Wills M, Lin R-B, Jones M. Do u
smoke after txt? Results of a randomised trial of smoking cessation using
mobile phone text messaging. Tob Control. 2005;14(4):255–61. https://
doi. org/ 10. 1136/ tc. 2005. 011577.
49. Cugelman B. Gamification: What it is and why it matters to digital health
behavior change developers. JMIR Serious Games. 2013;1(1):e3. https://
doi. org/ 10. 2196/ games. 3139.
50. Biello KB, Horvitz C, Mullin S, Mayer KH, Scott H, Coleman K, Dormitzer
J, Norelli J, Hightow-Weidman L, Sullivan P, Mimiaga MJ, Buchbinder S,
Bojan K, Futterman D, Emmanuel P, Liu A. HIV self-testing and STI self-
collection via mobile apps: experiences from two pilot randomized con-
trolled trials of young men who have sex with men. mHealth. 2021;7:26.
https:// doi. org/ 10. 21037/ mheal th- 20- 70.
51. Lee KJ, Carlin JB. Multiple imputation for missing data: Fully conditional
specification versus multivariate normal imputation. Am J Epidemiol.
2010;171(5):624–32. https:// doi. org/ 10. 1093/ aje/ kwp425.
52. Little R, Rubin D. Statistical Analysis with Missing Data. Vol 793. John Wiley
& Sons; 2019. https:// doi. org/ 10. 1002/ 97811 19482 260.
53. Pew Research Center. Teens, social media, & technology 2018. Published
2018. https:// www. pewin ternet. org/ 2018/ 05/ 31/ teens- social- media-
techn ology- 2018/. Accessed 2 June 2021.
54. Centers for Disease Control and Prevention. HIV Surveillance Report,
2016. Vol 28. U.S. Department of Health & Human Services; 2016. https://
www. cdc. gov/ hiv/ pdf/ libra ry/ repor ts/ surve illan ce/ cdc- hiv- surve illan ce-
report- 2016- vol- 28. pdf.
55. Ybarra ML, Summers Holtrop J, Prescott TL, Strong D. Process evaluation
of an mHealth program: Lessons learned from Stop My Smoking USA,
a text messaging-based smoking cessation program for young adults.
Patient Educ Couns. 2014;97(2):239–43. https:// doi. org/ 10. 1016/j. pec.
2014. 07. 009.
56. Rubin DB. Inference and missing data. Biometrika. 1976;63:581–92.
57. Leacy FP, Floyd S, Yates TA, White IR. Analyses of sensitivity to the missing-
at-random assumption using multiple imputation with delta adjustment:
application to a tuberculosis/HIV prevalence survey with incomplete
HIV-status data. Am J Epidemiol. 2017;185(4):304–15. https:// doi. org/ 10.
1093/ aje/ kww107.
58. Dubois LZ, Macapagal KR, Rivera Z, Prescott TL, Ybarra ML, Mustanski B. To
have sex or not to have sex? An online focus group study of sexual deci-
sion making among sexually experienced and inexperienced gay and
bisexual adolescent men. Arch Sex Behav. 2015;44(7):2027–40. https://
doi. org/ 10. 1007/ s10508- 015- 0521-5.
59. Ybarra ML, DuBois LZ, Parsons JT, Prescott TL, Mustanski B. Online focus
groups as an HIV prevention program for gay, bisexual, and queer adoles-
cent males. AIDS Educ Prev. 2014;26(6):554–64. https:// doi. org/ 10. 1521/
aeap. 2014. 26.6. 554.
60. White IR, Royston P, Wood AM. Multiple imputation using chained equa-
tions: Issues and guidance for practice. Stat Med. 2011;30(4):377–99.
https:// doi. org/ 10. 1002/ sim. 4067.
61. Fairchild AJ, Abara WE, Gottschall AC, Tein JY, Prinz RJ. Improving our ability
to evaluate underlying mechanisms of behavioral onset and other event
occurrence outcomes: a discrete-time survival mediation model. Eval Health
Prof. 2015;38(3):315–42. https:// doi. org/ 10. 1177/ 01632 78713 512124.
62. SRNT Subcommittee on Biochemical Verification. Biochemical verification
of tobacco use and cessation. Nicotine Tob Res. 2002;4(2):149–59. https://
doi. org/ 10. 1080/ 14622 20021 01235 81.
63. Fontanarosa P, Bauchner H, Flanagin A. Authorship and Team Science.
JAMA. 2017;318(24):2433–7. https:// doi. org/ 10. 1001/ jama. 2017. 19341.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub-
lished maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.