Objectives To inform the development of targeted sexually transmitted infection (STI) control programmes for persons who inject drugs (PWID).
Methods We recruited 116 PWID (aged ≥ 18 years) from a community-based syringe exchange programme (SEP) and assessed their STI knowledge and screening preferences via technology assisted self-interview. We estimated prevalence of STI transmission knowledge, attitudes and screening preferences as well as the association between reported sexual behaviours (past 6 months) and willingness to self-collect specimens.
Results Participants were white (77%), female (51%) and heterosexual (77%). STI knowledge regarding transmission and testing was high among the sample. More than 70% of participants were aware extragenital infections were possible and were least likely to know urine tests do not detect rectal infections (40.9%). Site-specific specimen collection was highly reflective of reported sexual behaviour. PWID who reported receptive sex (36% vs 5%, p<0.01) and insertive anal sex (31% vs 6%, p=0.01) were more likely to collect rectal specimens than those who did not. A similar trend was seen for oral sex performance on men and self-collection of oropharyngeal swabs (15% vs 3%, p=0.04). In addition, participants preferred collecting their own sample to having a clinician collect it for them (69% vs 31%, p<0.01) and testing at the SEP compared with a STI clinic (86% vs 14%, p<0.01).
Conclusion Our findings suggest site-specific specimen collection may be a proxy for risk behaviour engagement in this fairly knowledgeable high-risk population. To increase case finding, STI control programmes should educate patients about site-specific screening and pair outreach with the infrastructure provided by SEPs, in settings where these programmes exist.
- injecting drug use
- infection control
Statistics from Altmetric.com
Recent evidence suggests sexually transmitted infections (STI) are endemic to injection drug networks which raises questions about the importance of sexual transmission of HIV as a downstream effect of STI among this population.1–7 There is a growing body of evidence that suggests STI are a biomarker of imminent HIV infection among individuals in high-risk sexual networks.8–10 In New York City, 1 in 20 men who have sex with men (MSM) diagnosed with syphilis tested positive for HIV within a year.9 In Uganda and Zimbabwe, women diagnosed with trichomoniasis had 2.74 greater odds (95% CI 1.25 to 6.00) of acquiring HIV within 30 months.10 While evidence regarding screening efforts is limited among persons who inject drugs (PWID), studies of high-risk persons in STI prevalent networks indicate the importance of detection and treatment of STI as a strategy to reduce morbidity and HIV infection.
In the USA, few resources have been devoted to seeking, testing and treating STI among PWID likely because they are not considered a priority population for STI control according to current CDC testing and treatment guidelines.11 Correspondingly, there are no national surveillance data for this population12 and little research has focused on PWID preferences for accessing screening services within the USA or globally. An extensive literature search netted one study of sampling preferences among PWID accessing a street outreach clinic in Melbourne (Australia). In this sample, most preferred self-collecting specimen to practitioner-collected specimen (76% vs 9%; p<0.01).13 This finding is similar to other studies among highly vulnerable populations such as homeless youth,14 men-who-have-sex,15–17 and reproductive age women,18–23 all of which indicate self-sampling is preferable to clinician-sampling and that screening in outreach settings is preferable to those that are clinic-based.24–27
In a study conducted between July 2015 and February 2016, we documented high rates of STI among PWID recruited from a syringe exchange programme (SEP) in the Camden, New Jersey, USA.7 We offered multisite screening (genital, oropharyngeal and rectal) for chlamydia and gonorrhoea and participants collected specimen from each site based on their perceived risk. The positivity rate among women was 27% which was three-fold greater than men (8%).7Compared with the USA national prevalence, chlamydia estimates among our sample was approximately six-fold greater (eg, 1.7% in the NHANES sample from 2007 to 2012).28 In this paper, we present data not previously published regarding PWID’s STI knowledge, STI attitudes and STI screening behaviours/preferences in order to inform the development of outreach programmes targeted for PWID.
This study is part of a larger project to estimate STI prevalence among PWID which has been previously described.7 Briefly, participants were recruited from a SEP operating twice weekly in a small city (population: 75 000) in Southern New Jersey (USA). Eligible participants were English-speaking SEP clients who were at least 18 years of age, reported using injection drugs and engaging in oral, vaginal or anal sex within the past 6 months. All data were collected via technology-assisted self-interview on mobile tablets. Ethical approval for this study was provided by the Drexel University Institutional Review Board (protocol # 1506003729) and the Camden Area Health Education Centre Executive Director.
Sociodemographic characteristics, drug use and sexual behaviour
We collected information on sociodemographic characteristics (ie, age, sex at birth race/ethnicity, sexual orientation, education and income); sexual behaviour measures in the past 6 months that included performing or receiving oral sex (yes/no), vaginal sex (yes/no), insertive or receptive anal sex (yes/no), engagement in transactional sex (yes/no) and condom use during vaginal/anal intercourse (5-point Likert scale, ranging from never to always). Inconsistent condom use was operationalised by any participant not indicating ‘always’ using condoms.
Knowledge on STI transmission and screening was assessed using four true/false items. Items examining knowledge included, ‘it is possible to have an STD in your rectum (butt)/throat’ and ‘a urine test will detect if you have an STD in your rectum (butt)/throat’. Participant responses were assigned a score of 1 or 0 to indicate a correct or incorrect response, respectively.
Willingness and rationale for collecting extragenital specimen
Participants were asked two binary questions with yes/no response options to measure their willingness, a proxy measure for behavioural intention or perceived likelihood of engaging in extragenital sample collection. These included: ‘Today you are going to have an opportunity to have your rectum (butt) tested for an STD. Do you plan to test your butt today?’ and ‘Today you are going to have an opportunity to have your throat tested for an STD. Do you plan to test your throat today?’ When someone indicated they were unwilling to be tested in a specific site they received a follow-up question measuring reasons for not being willing to collect specimen at each site: ‘Why might you not consider getting your rectum (butt)/throat screened for STD?’ Participants then selected one response from: ‘I am not at risk’, ‘afraid it would hurt’, ‘afraid you might injure yourself’, ‘afraid it would be embarrassing’, ‘concerned it would be too messy’, ‘not willing to put anything in your rectum (butt)/throat’, ‘not comfortable with the setting’.
Previous and future engagement in STI screening
Participants were asked if they had received STI screening (yes/no) within the preceding 12 months by site: rectal and oropharyngeal. They were also asked four binary questions with yes/no response options to measure their future willingness engaging in extragenital sample collection. ‘Would you be willing to get tested for STIs again in the future?’; ‘Would you be willing to tell your friends that the Syringe Access Programme is providing free STI testing?’; ‘Would you be willing to take referral coupons to your friends? If they brought back the coupon, they could come to the syringe exchange van and get tested for free just like you did today?’; ‘If you test positive for an STI, would you be willing to take medication back to your sexual partner(s) so they could get treated as well?’
STI screening experience and preferences
STI screening experience and future screening preference was examined with the following questions: ‘How would you describe the experience of collecting your own sample to your friends?’ (answer choices were on a 5-pont Likert scale ranging from easy to difficult, painless to painful and comfortable to awkward); ‘ Next time you get tested, would you rather collect your own sample like you just did, or have a clinician collect it for you?’ (binary answer choices included, ‘I’d rather collect my own sample’ and ‘I’d rather have the clinician collect my sample’) and ‘Next time you get tested, would you rather be tested at the syringe exchange van, or would you prefer to go to a clinic (like your regular doctor or an STD clinic)?’ (binary answer choices included, ‘I’d rather get tested at the syringe exchange van’ and ‘I’d rather go to the clinic’).
Descriptive statistics were used to describe the sample as well as their STI knowledge, attitudes and preferences for future screening. χ² tests for equal proportions were used to detect differences between categories of preferred testing location and collection method. In addition, we conducted Pearson’s χ² tests or Fisher’s exact tests (for expected cell size <5) to assess the relationship between sexual behaviours and self-collection of genital, oropharyngeal and rectal specimens. Analyses were conducted using SPSS V.24 (IBM, Armonk, New York, USA).
A total of 138 PWID were recruited and screened for our study. Of these, 116 (84%) participants indicated sexual activity in the past 6 months and were retained for our final analyses. Sexual behaviours and risk factors among sexually active male (49%) and female (51%) injection drug users were previously reported.7 We described sociodemographic characteristics, STI transmission knowledge and attitudes towards screening of study participants (table 1). Among sexually active PWID, the median age was 32 (range 18, 62). Seventy-seven percent of participants were white, 77% self-identified as straight, 71% were homeless in the past 6 months, two-thirds completed high school or beyond and 75% had an annual income less than $9999.
The majority of participants were aware STI could be acquired in the oropharynx (75%) and rectum (79%) despite this 79% had not received a STI test in the preceding 12 months. Of those who had been screened, 11% reported collecting oropharyngeal specimens, 7% reported collecting rectal specimens and only 3% reported collecting specimens from both sites. Three-quarters of participants knew pharyngeal infections were not detectable via urine tests, but only 59% were aware that urine tests fail to identify rectal infections.
Nearly all participants (98%) reported they were willing to undergo STI screening in the future. Further, willingness to engage in peer-to-peer outreach was also high. Nearly all participants indicated they would be willing to refer a friend for screening (97%), take a friend referral coupons for free testing (96%) or bring presumptive treatment to partner if they were to screen positive (97%).
Most participants indicated self-collection of non-genital swabs was easy (87%), painless (89%) and comfortable (77%) (table 2). Participants preferred collecting their own sample to having a clinician collect it for them (69% vs 31%, p<0.01) and to testing at the SEP compared with a STI clinic (86% vs 14%, p<0.01). We did not collect data on willingness to collect genital specimen; however, 99% provided a urine sample at the time of the study visit. For those endorsing willingness to collect oropharyngeal (n=82) or rectal specimen (n=31), these samples were provided by 74% and 26% at the study visit. Of those who reported being unwilling, 31% and 3% also collected oral and rectal swabs, respectively (data not shown).
The main reason for declining both oropharyngeal and rectal screening was ‘no perceived risk’ of infection. Of those unwilling to collect oropharyngeal or rectal specimen, 58% and 45% reported no perceived risk as driving their intention, respectively. A smaller subgroup (11%) of the sample indicated discomfort with the setting and testing process (eg, afraid of self-injury).
Self-collection of oropharyngeal and rectal samples was significantly associated with risk behaviours (table 3). For example, collection of oral swabs was significantly associated with receiving (88% vs 47%, p<0.01) and performing (91% vs 41%, p<0.01) oral sex to men. Participants electing to not collect oral samples were more likely to receive oral sex from women (79% vs 46%, p<0.01). Vaginal sex was also significantly related to collection of oropharyngeal samples (71% vs 38%, p<0.01). Collection of rectal swabs was associated with receptive anal sex (36% vs 5%, p<0.01), insertive anal sex (31% vs 6%, p=0.01) and performing oral sex to a man (15% vs 3%, p=0.04).
Data shown here adds substantially to the currently limited information available about STI screening preferences among PWID. Most participants in this study were fairly knowledgeable about STI. More than 75% were aware of extragenital infections were possible and that a urine-based test would not detect an infection in these sites. Of all the knowledge questions, participants were least likely to know that a urine test would not detect a rectal infection (40.9%). Educating patients about site-specific testing may be important component of outreach to this population and may serve to increase uptake of extragenital screening, especially collection of rectal specimen.
Uptake of extragenital screening was much lower than genital screening. Only 64% and 9% of participants collected oropharyngeal and rectal specimen, respectively while 99% of participants collected genital specimens. To understand these differences, we assessed concordance in reports of risk behaviour and sample collection. Participants who reported performing oral sex on a man were more likely to collect an oropharyngeal specimen. However, there was no significant difference in collecting oropharyngeal specimen between those who reported performing oral sex on a woman compared to those who did not. Self-collection of rectal specimen was significantly higher among those engaging in both insertive and receptive anal sex. This finding is similar to other studies of high-risk groups, including behaviourally bisexual women and MSM where collection of rectal specimen was consistent with engagement in anal sex.15 29 30 It is important to note that studies exploring the relationship between extragenital sampling and sexual risk behaviours are limited. Our analysis adds to the growing literature indicating that screening behaviour is reflective of sexual behaviour.
A strength of the study is the linkage of sample collection behaviour with self-reported willingness to be screened at each site (ie, testing intentions). Some studies of behavioural intentions do not go on to follow participant engagement in the behaviour they endorsed. In this study, we were able to demonstrate that there is a disconnect between intention and screening behaviour, especially for extragenital screening. Uptake was fairly low for rectal (31% willing and 9% uptake) and oropharyngeal (76% willing and 64% uptake). One explanation is that perceived risk of infection moderates the relationship between willingness and uptake such that willingness decreases when one perceives no risk. Our sample is too small/underpowered to test out this hypothesis. However, as noted above, agreement between risk behaviour and sample collection was high suggesting some other factor is driving this effect. Further research with larger study populations is needed to tease apart the relationship between willingness and uptake. In particular, longitudinal research that assess for fluctuation in risk over time and its impact on screening behaviour may be warranted.
With regard to the development of outreach programmes for PWID, this sample preferred to access STI screening at the SEP as opposed to attending a STD clinic. They also preferred self-collecting specimen over clinician-collected specimen. This finding is similar to other studies of sampling preferences in other high-risk populations.13,15,25,29-30 (online supplementary W1). Given the potential for endemic infections in this population and their well-documented challenges to engagement with the healthcare system, STD control programmes should consider non-clinic-based programme to increase access to screening and thus case finding (online supplementary W2 and W3). Further, nearly all participants were willing to engage in expedited partner therapy (EPT). Limited EPT data exist for such a high-risk and mobile sample. In places with limited disease intervention services, engaging PWID to deliver presumptive therapy could have great cost savings while also increasing the number of persons treated (online supplementary W4 and W5).
The following limitations should be considered when interpreting our findings. This exploratory study had a small sample of PWID who were offered an opportunity to self-collect samples for STI on a mobile SEP outreach van. Due to the nature of our study (ie, predominately self-identified heterosexual and white PWID), results may not be generalisable to other PWID communities with intersectional identities (ie, persons of colour and MSM). MSM who inject drugs are at greater risk for STI (online supplementary W6–W8); however, our sample does not capture this high-risk group. Our sample is reflective of the approximately 3100 unique clients served by SEP during the recruitment period. There demographics are: 64% male, 70% White and 31 years or older (65%) (M. Chavis, email communication, April 2016). Further, many participants chose not to collect rectal specimen which may be attributable to the large proportion of persons who did not report anal sex in our sample. Further research is needed to examine whether perceived shame/stigma impacted reporting anal sex; this may be especially important for samples of persons engaging in transactional sex or bisexual men. Given the limited data on PWID screening preferences qualitative studies may be helpful for understanding how/why these factors impact screening attitudes and behaviour within various subpopulations of PWID. Finally, self-report and social desirability may further bias our findings. However, the use of technology-assisted self-interview was used to curb these effects.
Pairing control efforts with the infrastructure provided by community-based syringe exchange programmes, in settings where these programmes exist, may be an effective strategy to seek, test and treat PWID for STI. Instituting self-collected specimen as the standard of care for STI control programmes targeting PWID may serve to contain costs as simple written and verbal instructions can be delivered by non-clinical, outreach staff and this methodology appears to be preferable. To increase uptake, programmes targeting PWID should consider educating patients about site-specific STI including risk behaviours associated with transmission and site-specific screening in order to help participants better assess their own risk and encourage uptake of appropriate screening.
Among persons who inject drugs (PWID) attending a community-based syringe exchange programme (SEP) knowledge regarding extragenital infections and the importance of site-specific screening was high.
Specimen collection reflected reported sexual behaviours. PWID who reported anal sex and performing oral sex were more likely to self-collect rectal and oropharyngeal swabs, respectively.
Overwhelmingly, participants preferred collecting their own sample to having a clinician collect it for them and testing at the SEP compared with a STI clinic.
Providing PWID with information about site-specific screening and pairing outreach with SEPs, in settings where these programmes exist, could increase case finding.
The authors would like to thank the participants, staff at the Camden Area Health Education Center, the New Jersey Department of Health, Division of HIV/AIDS, TB and STD Services and our Research Assistants (Brenna Aumaier, MPH, Jesse Goldshear, MPH and Marisa Felsher, DrPH(c)) for their invaluable support in conducting this study.
Handling editor Adam Huw Bourne
Contributors AR designed the study and oversaw all aspects of the project from data collection through dissemination. NKT performed the statistical analyses. AR and NKT wrote the manuscript. BVDP contributed to study design and provided extensive comments and revision for the manuscript. MC contributed to study design, data collection and provided comments on the manuscript.
Funding This study was sponsored by New Jersey Department of Health, Division of HIV/AIDS, TB and STD Services as well as Community Driven Research Day; a joint effort of Drexel University, the University of Pennsylvania, Temple University and The Children’s Hospital of Philadelphia.
Competing interests None declared.
Patient consent Obtained.
Ethics approval Drexel University Institutional Review Board and the Camden Area Health Education Center Executive Director.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement There are no unpublished data available.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.