Internet recruitment for sexually transmitted infection screening among men who have sex with men in Eastern Europe

Abstract

Men who have sex with men (MSM) face negative health outcomes such as sexually transmitted infections (STIs) at disproportionate rates. Nonetheless, infections may be underestimated due to limited uptake in testing. To increase testing, screening interventions have been utilized in the past; however, some have resulted in limitations such as poor recruitment. To increase recruitment for screening of MSM in Estonia, two different recruitment strategies were examined. Recruitment was separated into two promotional periods: passive and active. Passive consisted of banners on gay-related sites, while active also placed banners on websites to the general public such as Facebook linked to specific thematic pages and users self-identifying as men. More men were recruited during the active period of five weeks (n = 134) than the passive period of 46 weeks (n = 126). Active promotion was so successful in that the number of home sampling kit orders far exceeded what was projected, forcing promotion to the general public to be closed after 13 days. Recruiting MSM through a combination of general public and gay-related websites and applications has the ability to quickly recruit for testing interventions. This method can recruit a large number in a short amount of time; therefore, a budget must be planned accordingly to support testing for all that participate.

Keywords

Europe men homosexual screening

Background

It is well documented that men who have sex with men (MSM) face negative health outcomes such as sexually transmitted infections (STIs) including human immunodeficiency virus (HIV) at disproportionate rates.^1–4 Despite MSM accounting for relatively small percentages of adult male populations throughout EU countries (0.96% in Slovakia to 3.40% in the United Kingdom) a large proportion of all incident STI cases are attributed to unprotected MSM sexual contact.^3,5 For instance, in 2015, there were 24,541 incident cases of syphilis, and in 2014, there were 66,413 incident cases of gonorrhea throughout the EU where almost two-thirds (63%) and almost half (44%) of infections were among MSM, respectively.^6,7 Additionally, among the 29,747 individuals diagnosed with HIV in 2015, 42% were MSM.⁸

Even though MSM are considered a key high-risk population for contracting STIs, infections among them may be largely undiagnosed in parts of the EU as there has been an insufficient uptake in testing, potentially due to societal climates such as stigma.⁹ To enhance testing, screening interventions have been used in the past.^10–12 Various field-based recruitment strategies include respondent-driven sampling (RDS), venue-based sampling, and street outreach.^12–14 However, to effectively recruit a large sample, these methods commonly rely on a well-networked community where similar individuals are socially connected with each other.^15–17 As a result, some screening interventions using field-based strategies have been limited by an inability to recruit adequate sample sizes.^18,19

As an alternative to field-based strategies, screening interventions have utilized web-based services as recruitment tools, specifically gay-related websites and applications.^20–22 Each have resulted in their respective challenges such as cost, retention, and participation.^23,24 Unfortunately, in some regions, exclusively using gay-related websites and applications limits the recruitment access to gay, bisexual, or other MSM who do not use them. For instance, engaging Estonian MSM in screening interventions via gay-related sites has been difficult due to their limited use.²⁵ The current project details an internet recruitment strategy providing free HIV/STI screening among MSM in Estonia. Throughout the country, there are approximately 9000 MSM with an estimated prevalence of HIV from 2% to 3%.^25–27 Options for HIV/STI testing in Estonia are limited as home sampling kits are available for STIs (with the exception of HIV) but they are not free. HIV testing must be completed in a clinical or community-based setting. Additionally, there are no permanent gay-friendly testing services available. To address these limitations, a general web-based recruitment screening intervention could be a viable approach in Estonia as 88% of households have internet access.²⁸ This project builds on previous interventions in other countries which indicate that MSM can be reached through general public social networking sites as well as those specific to the gay community.^29–31 The aim was to examine and compare the effectiveness of using general population social media sites to gay-related ones in recruiting for HIV/STI screening in Estonia.

Methods

Participants

Anonymous free HIV/STI testing was offered to MSM from February 2014 to February 2015 in two separate promotional periods: passive and active. Passive promotion was conducted from February 2014 through January 2015 where active was conducted from January 2015 through February 2015. Piloting and initial outcomes of these services in 2013 are described elsewhere.²⁵ Participants were linked to a web-based testing service to order home sampling kits as well as laboratory blood-based testing. Prior to placing orders, participants were required to indicate that they understood the aims and methods of the intervention and that they consented to take part in it. Additionally, each participant was requested to answer six questions including: age, gender, education, occupation, region of living and ever having sex with a man. Even though MSM were our target sample, all men were eligible to receive sampling kits regardless of reported MSM behavior in an attempt to reduce bias in answers provided for the six questions. Additionally, it would allow us to make comparisons between non-MSM and MSM through recruitment periods. Participants were provided a unique identifier to order tests and receive results. All data collected for ordering and delivery of testing kits were collected in a different database apart from the questionnaire. Ethical approval was not required for this study as it was considered a service evaluation and in Estonia, ethical approval is not required for service evaluations.

Promotion

All promotional materials were displayed in Estonian and Russian in both promotional periods: passive and active. Passive promotion spanned 46 weeks where banners linked participants to the opening page of the testing site. These were placed on gay- and HIV/STI-related Estonian websites throughout the project period. In addition, banners were displayed on PlanetRomeo’s homepage for two weeks. Messages read, ‘A healthy man does not need to hide his choices! Anonymous and free-of-charge sexual health check’.

Active promotion spanned five weeks where banners displayed on websites and apps including Facebook, Grindr, and GayRomeo read, ‘Who has sex, will test! Want to look good for your boyfriend?’ Grindr banners and broadcasts were displayed to geo-located users in Estonia. Banners on Facebook were displayed to males in the general public, including users linked to specific thematic pages and users self-identifying as men. Banners were also strategically placed on Google search windows and websites based on keyword searches using GoogleDisplay, GoogleAdwords, and SmartAD.

During active promotion, banners linked participants to a campaign website which was created to attractively provide accurate, culturally appropriate information devoid of medical jargon. Information included the importance of regular testing as well as testing locations. The campaign website directed participants to the testing site where they could order tests.

The total cost for home testing, including sample collection kits, testing of samples, postal costs, and the web-based testing service was 174.55 Euros/person. Blood-based testing cost an additional 21.84 Euros/person. The total budget for the campaign was 15,372 Euros. However, the bulk of the budget went towards the campaign website which was 10,532 Euros. The remaining 4480 Euros was used for advertising.

Testing

Several tests were available including gonorrhea, trichomoniasis, chlamydia, lymphogranuloma venereum (LGV), and mycoplasmosis. Kits included first void urine collection as well as anal and throat swabs. Directions for proper use and return were available in Estonian and Russian. Once kits were returned to the lab, test results were available within five business days and could be received through the testing website. Blood-based testing for HIV, hepatitis C virus (HCV), hepatitis B virus (HBV), hepatitis a virus (HAV), and syphilis was completed in person at one of six laboratory sites throughout the country. Test results were available within five business days and could only be received in person. Details on laboratory testing procedures are described elsewhere.²⁵ Personal information provided for blood-based testing were never linked to each participant’s unique identifier.

Participants testing positive for an STI through home sampling kits were provided treatment information including a location to receive free treatment (available only in capital city Tallinn). Participants testing positive for blood-based tests were recommended to seek confirmatory testing with an infectious disease doctor.

Data management and analysis

Data from the questionnaire and the unique identifier were stored in different databases and manually linked based on date and time of day the participant submitted the questionnaire and the date and time they reached the landing page. All personal information required for ordering test kits such as delivery address, phone number, and email were stored only on the testing database and never linked with questionnaire responses. All analyses were completed in Stata 13.1 (StataCorp LP. College Station, TX). Descriptive statistics were used to characterize participants.

Results

During the entire project, 436 men completed the questionnaire, 348 (78.8%) of them ordered sampling kits for testing where 260 (74.7%) were returned and processed. Out of the 260 men to test for STIs, 66% reported at least one sexual encounter with another man in his lifetime. Characteristics of all men (n = 436) who completed the questionnaire are presented in Table 1. Since non-MSM and MSM were both eligible to receive testing kits, data are presented for both. Sampling kit order information by promotion period is presented in Figure 1.

Table 1.

Characteristics of male participants, Estonia, 2014 (n = 436).^a

	Frequency (column %)		Frequency (column %)
	Passive period^b (n = 203)		Active period^c (n = 233)
Variable	Non-MSM (n = 73)	MSM (n = 130)	Non-MSM (n = 75)	MSM (n = 158)
Age (years)	Mean: 30.70; range: 17–68	Mean: 31.25; range: 14–65	Mean: 30.10; range: 20–67	Mean: 28.45; range: 16–55
Education
Secondary	6 (8)	5 (4)	11 (15)	14 (9)
High school	17 (23)	46 (35)	21 (28)	52 (33)
Vocational	15 (21)	24 (18)	24 (32)	24 (15)
College or more	35 (50)	55 (42)	19 (25)	68 (43)
Employment status
Employed	56 (77)	103 (80)	63 (84)	109 (69)
Unemployed	2 (3)	3 (2)	6 (8)	8 (5)
Student	14 (20)	16 (12)	5 (7)	36 (23)
Other	1 (1)	7 (6)	1 (1)	5 (3)
Language^d
Estonian	70 (96)	113 (87)	71 (95)	147 (93)
Russian	3(4)	17 (13)	4 (5)	11 (7)

MSM: men who have sex with men.

Numbers do not always add up to 436 due to missing values, only available data are reported. Percentages may not add up to 100 due to rounding.

28 February 2014–18 January 2015.

19 January–22 February 2015.

Language questionnaire was completed in is used as a proxy for language.

Figure 1.

Sampling kit order information by promotion period, Estonia 2015 (N = 348). *indicates the number of men with at least one positive STI test, does not reflect coinfection or infection at different anatomical sites. Does not include preliminary blood-based testing results.

More men were recruited for testing during the active promotion period of five weeks (n = 134) than the passive promotion period of 46 weeks (n = 126). Active promotion was so successful in that the number of home sampling kit orders far exceeded what was projected. Therefore, promotion to general internet sites including Facebook, GoogleDisplay, GoogleAdwords and SmartAD had to be closed after 13 days. Otherwise, funding would not have been able to support the number of tests if orders increased on the same trajectory. Thus, active promotion was only partially live. Throughout the entire campaign period, a total of 6769 IP addresses visited the campaign website 7496 times. Breakdown of visits indicates that 72% of the sessions started through campaign banners. Facebook banners were the most popular as 3.6% of those who saw them clicked on them. Nine percent were referred from internet forums, 2% by GoogleAdwords (the most popular were HIV, HIV testing, hepatitis, STIs, sex between, gay, and homo) and 1% through social media. Of advertisement routes with engagement data, GoogleAdwords banners were clicked 717 times, SmartAD banners were clicked 3086 times and Facebook banners were clicked 2352 times. Based on total impression engagement, Facebook banners were the most popular.

The vast majority of kits were returned in one month (74.1%) or two months (94.1%) from order time. During passive promotion, 167 kit orders were made with 41 unreturned resulting in a 75% return rate. During active promotion, 181 kit orders were made with 47 unreturned resulting in a 74% return rate. A total of 11.9% (n = 31) of participants had some STI diagnoses from urine and/or anal and/or throat swabs. The proportion of men testing positive for any STI decreased to almost half when comparing passive and active promotion periods, 15.9% to 8.2%, respectively.

During both the passive and active recruitment periods, STI diagnoses were slightly higher in MSM vs. non-MSM; however, neither reached statistical significance. In the passive promotion period, out of the 44 non-MSM, 4 tested positive for an STI whereas 16 out of the 82 MSM tested positive (χ²= 2.33, p = 0.13). In the active recruitment period, among the 44 non-MSM, 2 tested positive for an STI whereas 9 out of the 90 MSM tested positive (χ²= 1.17, p = 0.28).

Discussion

The results indicate that strategies used in active promotion of HIV/STI testing to the general population using internet sites like Facebook have the potential to increase STI testing uptake in this high-risk population. Advertising on Estonian gay-related websites during the passive promotion period may have resulted in poor recruitment as these sites are more focused on gay rights and individuals may not visit them frequently. Previous research indicates that web-based campaigns to promote home STI testing to MSM not only increases access to a viable method of STI testing,²¹ but also is potentially more cost effective than clinic-based testing.²² Even though our costs to detect a positive STI test tripled in the active promotion period, short-term investments in advertising have the potential to pay off by recruiting more men. From a public health perspective, this strategy has the ability to prevent transmission and subsequent costs resulting in long-term health and economic benefits. Also, broader exposure to promotional materials enforces the importance of testing and has the ability to identify infections that otherwise may have remained undiagnosed.²⁰

Like the majority of previous studies in Estonia, the proportion of Russian speakers was low with less than 10% completing the questionnaire in Russian.²⁵ This is of particular importance as 25.1% of the Estonian population are Russian.³² Additionally, the majority of participants reported living within urban municipalities. This finding is consistent with previous research in not only Estonia but also other Eastern European countries such as Latvia and Lithuania. For instance, results from a large scale internet study among MSM in 38 European countries indicated that Lithuania and Latvia had high percentages of respondents living in urban rather than rural settings.^5,33 Continuously low percentages may reflect that there are more MSM residing in urban municipalities, or it may highlight an inability to reach those in rural ones.

We deliberately offered free testing to all men in order to ensure reliable questionnaire responses. For instance, we did not want men reporting no lifetime MSM behavior to retake the questionnaire falsely identifying as MSM if they were initially turned away. The proportion of MSM recruited during both promotion periods was similar, 65.1% in the passive and 67.2% the active, indicating that the active promotion approach was effective in reaching our target group. Interestingly, 35.0% of sampling kit orders were placed by non-MSM during the passive promotion; yet, the majority of recruitment materials were displayed on gay-related websites and applications. We expected that more non-MSM would be recruited through active promotion as most materials were displayed on sites to the general public, yet only 32.8% of orders were placed by non-MSM. As this was a self-reported behavior, percentages may reflect that men were unwilling to identify as MSM or had not participated in sexual contact with another male.

The proportion of men testing positive for an STI decreased almost twofold when comparing passive and active promotion periods. Not only could this difference be attributed to the different sites used in each promotion period it also could be related to a sense of urgency for testing as the ordering period was much shorter during active promotion. For instance, passive promotion banners did not include an order deadline where active promotion banners did. Placing an order deadline may have motivated the ‘worried well' to order kits which would inflate the proportion of negative tests.

Limitations

There were multiple limitations, including the eligibility criteria. As all questionnaire answers were self-reported, women could have ordered testing kits. The word ‘sex’ was not defined when asking about sexual behavior with other men as it was intentionally left open to interpretation. This could have decreased the number of people reporting a sexual history with other men as some do not view oral sex or even insertive practices as ‘sex’.

As we did not save IP addresses for anonymity reasons, we were unable to track the amount of people clicking specific banners who ordered tests and what the proportion of MSM was among them. We know the number of Facebook clicks that landed on the campaign website, but we do not know how many of them proceeded. We cannot determine which sites were more effective in increasing the number of people who actually got tested. We can only say that the combination of venues we used was successful in reaching more men. However, ensuring anonymity could also be a strength as it would not discourage an individual from participating. Due to the anonymous nature of the study, we were also unable to identify the number of participants who accessed confirmatory testing for positive blood-based screening results. Also, we were unable to track whether or not those who screened positive were linked to care.

The use of GoogleAdwords utilized very classical and traditional search words in both Estonian and Russian. Future work can include a more detailed analysis to determine if there were potentially more appropriate words in each language.

Conclusions

Despite limitations, MSM recruitment through general population websites has the potential to recruit those who may not visit gay-related sites and social networking apps. Investing in advertisement methods utilized during the active promotion period increased testing, reinforced the importance of testing, and linked persons to care. As such, this method can potentially recruit a large number of people in a short amount of time, and therefore a budget must be planned accordingly to support testing for all that participate. In the future, we suggest to request more detailed feedback from those who access services and consider strategies to retain participants from testing to linkage in care.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work has been supported by the National Institute for Health Development, Estonia, from Estonian Research Council Health promotion research programme TerVE (grant number 3.2.1002.11-0002) and National Health Plan for 2009–2020.

References

Beyrer

Baral

van Griensven

et al.

Global epidemiology of HIV infection in men who have sex with men. Lancet 2012; 380: 367–377.

CDC. Today's HIV/AIDS epidemic, Atlanta, GA: Center for Disease Control and Prevention, 2015.

ECDC. Annual epidemiological report 2014 – sexually transmitted infections, including HIV and blood-borne viruses, Stockholm: European Centre for Disease Prevention and Control, 2015.

WHO. Global report: UNAIDS report on the global AIDS epidemic 2012, Geneva: World Health Organization, 2012.

Marcus

Hickson

Weatherburn

et al.

Estimating the size of the MSM populations for 38 European countries by calculating the survey-surveillance discrepancies (SSD) between self-reported new HIV diagnoses from the European MSM internet survey (EMIS) and surveillance-reported HIV diagnoses among MSM in 2009. BMC Public Health 2013; 13: 919–919.

ECDC. Annual epidemiologic report 2016 – Gonorrhoea, Stockholm: ECDC, 2016.

ECDC. Annual epidemiological report 2016 – syphilis. Stockholm: European Centre for Disease Prevention and Control, 2016.

ECDC. HIV/AIDS surveillance in Europe 2015. In: Stockholm: European Centre for Disease Prevention and Control, 2016.

ECDC. HIV and STI prevention among men who have sex with men, Stockholm: European Center for Disease Prevention and Control, 2015.

10.

Katz

Cassels

Stekler

. Replacing clinic-based tests with home-use tests may increase HIV prevalence among Seattle men who have sex with men: evidence from a mathematical model. Sex Transm Dis 2014; 41: 2–9.

11.

Rife

Cate

Kosinski

et al.

Participant recruitment and data collection through Facebook: the role of personality factors. Int J Soc Res Methodol 2016; 19: 69–83.

12.

Soni

White

. Self-screening for Neisseria gonorrhoea and Chlamydia trachomatis in the human immunodeficiency virus clinic – high yields and high acceptability. Sex Transm Dis 2011; 38: 1107–1109.

13.

Gold

Pedrana

Sacks-Davis

et al.

A systematic examination of the use of online social networking sites for sexual health promotion. BMC Public Health 2011; 11: 583–583.

14.

Mansergh

Koblin

McKirnan

et al.

An intervention to reduce HIV risk behavior of substance-using men who have sex with men: a two-group randomized trial with a nonrandomized third group. PLoS Med 2010; 7: e1000329–e1000329.

15.

Uusküla

Johnston

Raag

et al.

Evaluating recruitment among female sex workers and injecting drug users at risk for HIV using respondent-driven sampling in Estonia. J Urban Health 2010; 87: 304–317.

16.

Heckathorn

. Snowball versus respondent-driven sampling. Soc Methodol 2011; 41: 355–366.

17.

McCreesh

Frost

Seeley

et al.

Evaluation of respondent-driven sampling. Epidemiology 2012; 23: 138–147.

18.

Kuhns

Kwon

Ryan

et al.

Evaluation of respondent-driven sampling in a study of urban young men who have sex with men. J Urban Health 2015; 92: 151–167.

19.

Rüütel

Trummal

Salekesin

et al.

HIV epidemic in Estonia: analysis of strategic information, Geneva: World Health Organization, 2011.

20.

Elliot

Rossi

McCormack

et al.

Identifying undiagnosed HIV in men who have sex with men (MSM) by offering HIV home sampling via online gay social media: a service evaluation. Sex Transm Infect 2016; 92: 470–473.

21.

Wood

Ellks

Grobicki

. Outreach sexual infection screening and postal tests in men who have sex with men: are they comparable to clinic screening? Int J STD AIDS 2015; 26: 428–431.

22.

Blake

Spielberg

Levy

et al.

Could home sexually transmitted infection specimen collection with e-prescription be a cost-effective strategy for clinical trials and clinical care?

Sex Transm Dis 2015; 42: 13–19.

23.

Noble

Jones

Bowles

et al.

HIV testing among internet-using MSM in the United States: systematic review. AIDS Behav 2017; 21: 561–575.

24.

Lane

Armin

Gordon

. Online recruitment methods for web-based and mobile health studies: a review of the literature. J Med Internet Res 2015; 17: e183–e183.

25.

Rüütel

Lohmus

Janes

. Internet-based recruitment system for HIV and STI screening for men who have sex with men in Estonia, 2013: analysis of preliminary outcomes. Euro Surveill 2015; 20: 16–25.

26.

ECDC. Sexually transmitted infections in Europe 2012, Stockholm: European Centre for Disease Prevention and Control, 2014.

27.

Marcus

Schmidt

Hamouda

et al.

Estimating the regional distribution of men who have sex with men (MSM) based on internet surveys. BMC Public Health 2009; 9: 180–180.

28.

Statistics Estonia. Statistical Database. 2017. Available at: https://www.stat.ee/29992 (accessed 1 August 2017).

29.

Capurro

Cole

Echavarria

et al.

The use of social networking sites for public health practice and research: a systematic review. J Med Internet Res 2014; 16: e79–e79.

30.

Charles-Smith

Reynolds

Cameron

et al.

Using social media for actionable disease surveillance and outbreak management: a systematic literature review. PloS One 2015; 10: e0139701–e0139701.

31.

Sullivan

Khosropour

Luisi

et al.

Bias in online recruitment and retention of racial and ethnic minority men who have sex with men. J Med Internet Res 2011; 13: e38–e38.

32.

Statistics Estonia. Statistical database. 2017. Available at: https://www.stat.ee/34278?highlight=russian,population (accessed 24 July 2017).

33.

ECDC. The European men-who-have-sex-with- men internet survey. Findings from 38 countries, Stockholm: European Centre for Disease Prevention and Control, 2013.