Scientific Sessions 2015: IBBS survey 2014

Dr Sriyakanthi Beneragama, Consultant Epidemiologist

1. Integrated Biological and Behavioral Surveillance (IBBS) Survey 2014 Highlights Dr.Sriyakanthi Beneragama MBBS,MSc,MD ( Com.Med .),MIRB( A ustralia) Consultant Epidemiologist, Co- ordinator Special Surveys/Global F und Project, National STD/AIDS Control P rogrammme ,

2. HSS since – 1990 BSS - in 2006/2007 First IBBs – in 2014

3. Presentation Outline HIV s urveillance in different epidemic scenarios Objectives and Case Definitions Methodology Key findings Limitations Conclusions and Recommendations

4. Different Epidemic Scenarios

5. Different Epidemic Scenarios

6. HIV Surveillance Modalities in Low Prevalence settings

7. Objectives To estimate the prevalence of HIV, syphilis and associated risk behaviours amongst four key populations in Sri Lanka Female sex workers (FSW) Men who have sex with men (MSM) People who inject drugs (PWID) Beach boys (BB) To assess the use of and access to health and social welfare programs among FSW, MSM, PWID and BB in Sri Lanka

8. Definitions high risk groups- IBBS Female Sex Worker (FSW): Any female, who has sold sex in exchange of money or goods in the past six months. Men who have Sex with Men (MSM): Men who have had sex with another men in the past six months as a matter of preference or practice, regardless of their sexual identity or sexual orientation, and irrespective of whether they also have sex with women or not.

9. Definitions Contd.. People Who Inject Drugs (PWID): A person who have been injecting drugs during the 12 months preceding the study. Beach Boy (BB): Males who cruise in and around beach areas, and associate with tourists as guides, animators or providers of any form of gratification including insertive and receptive sex (homosexual, heterosexual or bisexual orientation), during the previous 12 months.

10. National estimates of HRGs Category Estimate (average) Range (95%CI) FSW 14,132 12,329 – 15,935 MSM 7,551 6,547 - 8,554 DU 17,459 15,338 - 19,542 IDU 423 328 - 516 BB 1,314 1,142 - 1,486 Source: National Size Estimation Survey. 2013

11. Sampling Methods-Hidden Populations Snow Ball Time Location Cluster (TLS ) Respondent Driven (RDS ) Non probability sampling Seeds selection give rise to selection bias in the final sample Probability sampling Assumption is MARP groups gather at locations Probability sampling Seed selection specific &does not require sample frame

12. Components of RDS

13. Methodology, Study locations and sample sizes District FSW MSM PWID BB Total Colombo 600 500 250   1,350 Kandy 350     350 Galle 300 350   300 950 Anuradhapura 350   350 Total 1,250 1,200 250 300 3,000

14. Interviewing and Electronic Data C ollection 17 October 2015 IBBS Sri Lanka 14

15. Open Data Kit (ODK) Open Data Kit (ODK) is a free and open-source set of tools which help organizations author, field, and manage mobile data collection solutions.  https://opendatakit.org/

16. Behavioral Component Recruiter and coupon information Demographic Characteristics Sexual History: Numbers & Types of Partners Sexual History: Transactional Sexual Intercourse Sexual History: Non Transactional / non-Paying Sexual Intercourse Male Condoms Female Condoms

17. Behavioral Component contd.. STIs and STI Treatment Seeking Behaviour Perception of HIV and STIs transmission risk HIV/AIDS Knowledge, Opinions and Attitudes (UNAIDS/GARPR indicators) Drug and Substance Use (including needle/syringe sharing) Media & HIV Intervention Programme Exposure Stigma & Discrimination

18. Biological testing component

19. HIV Testing Algorithm

20. Ethical Considerations Informed consent Voluntary Participation, If changed the mind option of dissent, clear understanding about the research Confidentiality Risk/Benefit Ratio Early identification of the disease status , Referral to care, knowledge about HIV/STI IGP was made aware and permission obtained , Data collection sites – client friendly environment Beneficence K nowledge, free condoms , HIV status and Syphilis testing Financial Incentive

21. RDSAT The Respondent Driven Sampling Analysis Tool (RDSAT) gives you estimation of important population parameters (population proportions of user defined groups, average personal network sizes, homophile , measures of statistical significance for population estimates ), data display and editing , export of analysis results to formatted html, data visualization and graphics.

22. Results: Sample diagnostics Sample size: 3,110 respondents, across 8 KAP/sites FSW Colombo: 605 FSW Galle: 302 FSW Kandy: 354 MSM Colombo: 504 MSM Galle: 355 MSM Anuradhapura:358 BB Galle: 306 PWID Colombo: 326 Maximum number of waves: 14 (BB Galle) Minimum number of waves: 8 (FSW Kandy)

23. Results: FSWs Characteristic Colombo Galle Kandy Mean Age & Range 38.8 years 18-72 37.4 years 20-75 34.6 years 19-65 Mean number of sexual partners during last 7 days & Range 6.7 0-35 6 0-20 6.9 0-60 Mean number of clients during the last day & Range 2.17 1-6 1.92 1-6 1.97 1-20 Consistent condom use 83.3% 75.4% 76.5%

24. Results: FSW Cont … Characteristic Overall prevalence/frequency estimate % SE* 95% CI** HIV Prevalence (Colombo and Galle) 1 0.0035 1.03 HIV Prevalence (All) 0.81 - - Syphilis - Active (All ) 0.98 - - Composite knowledge 34.93 0.0145 32.1 – 37.8 Used a condom at last sex with a client 92.96 0.0079 91.4 – 94.5 Tested for HIV in the past 12 months and knows result 35.01 0.0154 32.0 – 35.0 Received free condoms from NGOs or a health care center in the last 12 months 37.67 0.016 34.5 – 40.8 Reached with HIV prevention programs (received free condoms and know where HIV testing can be obtained) 29.85 0.0148 26.9 – 32.8

25. Results: MSM Characteristic Colombo Galle Kandy Mean Age & Range 34.3 years 18-71 23.1 years 18-55 30 years 18-65 Mean number of sexual partners during last 7 days & Range 2.2 0-45 1.7 0-20 1.4 0-15 Consistent condom use 41.8% 33.6% 35.2%

26. Results: MSM Cont.. Characteristic Overall prevalence/frequency estimate % SE* 95% CI** HIV Prevalence (Colombo and Galle) 1.03 0.0043 0.2 – 1.9 HIV Prevalence (All) 0.88 - - Syphilis - Active (All ) 1.96 0.0083 0.3 – 3.6 Composite knowledge 30.46 0.0162 27.3 – 33.6 Used a condom at last sex with a client 57.94 0.0197 54.1 – 61.8 Tested for HIV in the past 12 months and knows result 15.42 0.0169 12.1 – 18.7 Received free condoms from NGOs or a health care center in the last 12 months 26.85 0.0183 23.3 - 30.4 Reached with HIV prevention programs (received free condoms and know where HIV testing can be obtained) 19.29 0.0177 15.8 – 22.8

27. Results: BB and PWID Characteristic Galle (BB) Colombo (PWID) Mean Age & Range 27 years 18-56 40.2 years 18-66 Mean number of sexual partners during last 7 days & Range 0.9 0-15 2.7 0-50 Consistent condom use 69.8% With a tourist 68.1% 25.9%

28. Results: PWID Characteristic Overall prevalence/frequency estimate % 95% CI HIV -- Syphilis -- Composite knowledge 33.3 27.5 – 39.1 Used a condom with last sex sexual partner   24.0 18.8 – 28.9 Tested for HIV in the past 12 months and knows result   8.7   5.8 – 11.8 Received free condoms from NGOs or a health care center in the last 12 months   6.0 3.5 – 8.5 Reached with HIV prevention programs (received free condoms and know where HIV testing can be obtained) 4.1 2.2 – 6.1 Safe injecting practices Shared injecting equipment at last injection 43.5 43.5 – 55.0

29. Results: BB Characteristic Overall prevalence/frequency estimate % 95% CI HIV -- Syphilis -- Composite knowledge 20.1 16.0 – 23.4 Used a condom at last sex with a tourist 67.6 61.8 – 73.3 Tested for HIV in the past 12 months and knows result 4.3 1.5-7.4 Received free condoms from NGOs or a health care center in the last 12 months 15.1 10.9 – 18.7 Reached with HIV prevention programs (received free condoms and know where HIV testing can be obtained) 7.8 5.0 – 10.0

30. Limitations Network size Large number of ‘don’t know’ and ‘refuse to answer’ around network size questions New independent network size form introduced allowing two independent measures of network size which could be reviewed Lesson - Importance of training interviewers specifically on how to ask this question to limit missing data. Most noticeable for the PWID group, hence the interpretations of these findings are limited

31. Limitations contd.. Recruitment ‘Mobile sites’ to brothels and massage parlours introduced, potential for disruption of principles of natural peer recruitment

32. Limitations contd.. 3 . Population size estimations (PSE) As part of RDS-A PSE estimates must be entered into the software to enable population proportions to be generated Consensus amongst stakeholders is that current national PSE are underestimated across all groups and districts, particularly amongst PWID MF, KIT together with the SAC came up with revised estimates to enter into RDS-A .

33. Conclusions - FSW Overall Condom usage high at last sex with client, all other GARPR indicators (testing, knowledge, and prevention programmes) are low Entry into sex work Financial Abandoned by husband, Average amount money earned per last sex act between 1,276 Rupees (Colombo) and 1,822 (Kandy)= financial incentive

34. Conclusions - FSW contd.. Characteristics of sex work Three wheel / truck stops commonly reported place where FSW meet clients, potential for implementing interventions in truck stops / areas of mobility, such as mobile clinics, IEC, etc. Most common location where sex is being exchanged is hotels, providing opportunity for engaging with the private sector in a condom distribution strategy

35. Conclusions – FSW contd.. Male condoms High number of FSW have been paid more for sex with no condom, interventions targeting clients to raise awareness needed, and condom negotiation skills building Most condoms are currently obtained from private pharmacies, justifying engagement with the private sector, potential for a public/private partnership for condom distribution

36. Conclusions – MSM Overall Poor performance on all GARPR indicators, including condom usage, testing knowledge and prevention programmes Characteristics of sexual behaviour No evidence to support high presence of intergenerational sex, despite anecdotal evidence and discussions in various meetings and forums

37. Conclusions – MSM contd.. Sexual behaviour Anal sex is high amongst MSM across all districts, alongside poor condom usage, clearly providing a pathway for the increased spread of HIV. Presence of sex with women present, illustrating the potential for cross-spread between KAP and general populations Most MSM do not ask HIV status of last partner Sex work Many MSM have both sold and paid for sex - identities and boundaries sometimes blurred between MSM and sex work

38. Conclusions – MSM contd.. Condom usage Condom usage is low, approximately half of MSM using condoms at last sex with partner Condom usage at last sex with female appears to be lower than with males, this phenomenon could be explored further, and if necessary incorporated into the IEC and BCC materials Condoms not being available is a common reason for lack of condom usage, current condom distribution strategies should be reviewed for comprehensiveness Blurred sexual identities that are not well understood by service providers.  

39. Conclusions – MSM contd.. Stigma and Discrimination High levels of stigmatization in society, including in the family of MSM, leading to a non-conducive environment for safer sexual practices and access to health services Sexual Identifies Blurred sexual identities creating potential for multiple environments of risk, which are also not well understood by service providers and should be taken into account for successful prevention and treatment and care programmes

40. Conclusions – PWID Overall Poor performance on all GARPR indicators, including condom usage, testing knowledge and prevention programmes Characteristics of injecting drug behaviour Most PWID have been injecting for between six to 20 years Frequency of injecting is not high, with most PWID only injecting two to four times a month, once a month, or less, which may be influenced by the shortage of heroin at the time of the survey. Heroin is the most commonly used drug during injections.

41. Conclusions – PWID contd.. Sex work Low prevalence of selling sex, including selling sex to obtain drugs; however over a quarter have given money, goods or services to have sex with a partner Condom usage Condom usage is low, with typically less a quarter of PWID using a condom at last sex with a partner A common reason for not using condoms is ‘never heard of them’, indicating the need for increased education and awareness

42. Conclusions – PWID contd.. Safe injecting practices Substitution treatment and needle and syringe exchange programmes focused on enhancing safe injection practices seem absent A considerable percentage of PWID share their injection equipment Not all PWID have a clear understanding and or misconceptions about safe injection practices There is no information on the size of the sharing network itself between PWID Unsafe disposal of injecting equipment (needles and syringes) highly present, with most simply throwing them in the trash or in to open water, awareness campaigns around proper disposal are urgently needed

43. Conclusions – BB Overall Poor performance on all GARPR indicators, including condom usage, testing knowledge and prevention programmes Nature of interaction with tourists Most BB act as tour guides or hotel related staff More than a third of BB have sex with tourists Frequency of engaging in sex with tourists varies, with just under half indicating ‘sometimes’, and only just over 10% indicating ‘every time’ More than a quarter received money the last time they had sex with a tourist Most of the tourists are European

44. Conclusions – BB contd.. Sex work Nearly one third have received money, gifts, or favours in exchange for sex, and nearly a quarter have ever paid for sex Condom usage Over two thirds used a condom last time they engaged in sex with a tourist

45. Recommendations Increase condom awareness and usage Increase HIV testing Increase participation to address HIV Reduce stigma around key populations and HIV Innovate with HIV interventions Operational research needs

46. 8. Recommendations 3. Increase participation to address HIV Documented approaches on key population engagement in similar contexts should be explored and piloted in the Sri Lankan context (e.g. KAP led outreach, development of new and continuation of existing of FSW, MSM and other KAP collectives and social groups, and strengthening of community systems) These new approaches should be comprehensively evaluated with successes and continual gaps documented Specifically, an expanded harm reduction approach for PWID to address misconceptions around safer injection practices, as well as more discussion around substitution treatment is recommended 17 October 2015 IBBS Sri Lanka 47

47. 8. Recommendations 2. Increase HIV testing Review current peer educator model to assess strengths, weaknesses, impact and areas for improvement given the limited number of key populations who have been in contact with peer educators to date Develop tailor made BCC strategies and models for key population sub groups (strata ) Feasibility and acceptability of expansion of HIV rapid testing across STD clinics, in order to reduce waiting times and improve client experience, and ultimately uptake, should be explored Expansion of testing to additional venues is needed, for example moonlight HTC, outreach and mobile HTC General health practice testing including provider initiated testing and counselling (PITC) should be reviewed and explored 17 October 2015 IBBS Sri Lanka 48

48. 8. Recommendations 1. Increase condom awareness and usage Formulation and implementation of a condom policy / strategy national plan, to include Expansion of condom promotion and distribution through expanded channels Private sector collaboration, including pharmacies, hotels, around parks and public bathrooms and other venues Special attention given to enhancing condom use in difficult situations, i.e. in parks, public bath rooms, or when having unplanned sexual contacts Activities to reduce police arrest on the basis of carrying condoms   Interventions with clients of sex workers to carry and use condoms Creating an enabling environment through ongoing advocacy for access to condoms 17 October 2015 IBBS Sri Lanka 49