Drug situation in Greece

16/6/217 INDICATIONS FOR DECLINE IN HCV AND HIV INCIDENCE AMONG PEOPLE WHO INJECT DRUGS IN ATHENS, GREECE Anastasios Fotiou Greek Reitox focal point at the Athens University Mental Health Research Institute Collaboratively with Eleftheria Kanavou & Argyro Andaraki EMCDDA - DRID National Expert Meeting Session 4. Directions for the development of the DRID indicator Lisbon, EMCDDA, 15-16 June 217 Drug situation in Greece 2.4per 1 people (~17) aged 18-64use opioids ~1/3 current injectors and more than half live in Athens ~5 people enter treatment / OST) annually ~3/5 have entered treatment before Opioids responsible for ~3/4 of treatment entries The majority of entries in drug-free settings long and stagnant waiting list for entry in OST in Athens >6% OST coverage nationwide, just above 5% in Athens ~2% of current injectors report also current sharing of syringes ~19 syringes per current injector a year (215), scant in other cities ~63% HCV antibody prevalence among entries in Athens, higher in OST ~14% HIV prevalence among treatment entrants in Athens, higher among active users (~17%, 213 data) 1

16/6/217 Sharp increases in the yearly number of HIV notifications associated with IDU in 211 and 212 almost all reported in Athens 12 1 8 HIV infections associated with injection drug use, N 8 % of HIV infections associated with IDU (among cases with known mode of tranmission) 6 54 Number, Ν 6 4 33 551 36 4 Percentage, % 2 266 272 18 13 13 2 4 3 2 3 3 17 11 11 15 16 16 7 58 26 27 28 29 21 211 212 213 214 215 216¹ Fig. Trends in the number and proportion of HIV notifications associated with injection drug use in Greece Source: Hellenic Centre of Disease Control & Prevention Notes.¹ January-October 216 Key response #1: Scaling up of the syringe programs in Athens following the 211 outbreak Syringes distributed / exchanged, N 5, 45, 4, 35, 3, 25, 2, 15, 1, 5, 29782 25 (3537) Syringes exchanged/distributed, N Syringes per estimated injection* drug user, Ν 64958 3489 8 11 26 (3243) 5519 68579 61516 2 19 21 27 (326) 28 (2855) 29 (3194) 119397 16 21 (391) 46898 43 211 (283) 133 435617 212 (369) 223 31576 213 (1956) 139 26761 19 Source: Greek Reitox Focal Point of the EMCDDA, 216 214 (2272) 215 (2451) 25 2 15 1 5 Syringes per current injection drug user, N * Past 3 days injection Reporting year (Estimated number of current injection drug users; central estimate) Fig. Trends in the number of syringes distributed and the number of syringes per estimated current PWID in Athens 2

16/6/217 Key response #2: Increase availability of OST coverage 5 1 Number of opioid users on waiting list for entry in OST in Athens 4 3 2 1 3771 3838 3689 3569 3157 2757 2828 2431 47 51 43 33 231 32 28 181 22 22 22 25 26 27 28 29 21 211 212 213 214 9 8 7 6 5 4 3 2 1 Percentage of estimated problem opiod users in OST Fig. Trends in the number of opioid users on the waiting list to enter OST and percent of estimated problem opioid users in OST in Athens (coverage) Source: Key response #3: Seek, test, link to care, and prevent further harm The ARISTOTLE program Aim Seek, test, alert, link to care (as a response to the HIV outbreak) Site Athens (metropolitan area) Period Aug 212-Dec 213 (18 months) Setting Community Population PWID in the past 12 months, 18 years of age Method Respondent driven sample, RDS; 5 rounds; incentives Sample N=3,32 unique participants ( 36yo); 54% in 1 rounds Coverage 71-1% (based on HRDU estimates for current injectors) Data collection Blood sample and CAPI Source Hatzakis et al Addiction, 215 Run by Athens University Medical School, in collaboration with the Greek Organisation Against Drugs (OST), the Hellenic Centre for Disease Control & Prevention, and NGOs 3

16/6/217 ARISTOTLE reported significant changes towards less risky behaviours from round 1 to round 5 Decrease in daily injecting, from 45% to 19% Decrease in current sharing of syringes, from 37% to 26% Entry in OST doubled, from 12% to 28% HIV test uptake increased from 5% to 87% % of undiagnosed HIV-positives declined from 84% to 15% 48% of previously unlinked seropositives visited an infectious diseases unit, and one-fourth (25%) initiated ART after their participation and by the end of round 5 Source: Hatzakis et al Addiction 215; Sypsa et al J Inf Dis 217 ARISTOTLE enriched data on HIV prevalence by including PWID from wider networks in Athens In ARISTOTLE The prevalence of HIV infection ranged across rounds from 12% to 16% from 6% (last round) to 26% (first round) among new injectors (not shown) Relatively higher prevalence compared to treatment settings in general, but lower compared to OST (%) 4 35 3 25 2 15 1 5 PWID tested HIV positive in ARISTOTLE 212-213 (%) HIV positive PWID entering all types of treatment in Athens (%) HIV positive entering OST in Athens (%) 8 8 12 11 16.5% Fig. Prevalence of HIV infection among PWID entering (a) any type of drugrelated services in Athens, (b) OST and (c) ARISTOTLE program Source: ; Hatzakis et al Addiction 215; Sypsa et al J Inf Dis 217 18 14 15 11 35 14 9 13 216 data are preliminary, not to be cited 28 29 21 211 212 213 214 215 216 4

16/6/217 ARISTOTLE data allowed the direct measurement of HIV incidence among PWID participating in the program Infectious disease: HIV Sample / data: PWID in the past 12 months Participants of the 5-round RDS of the ARISTOTLE program N = 1521 with >1 presentation in the program Period: 212-213 Type of measurement: Direct, longitudinal follow up of PWID who where initially negatives and seroconverted during the program Seroconversion time: Mid-point between last negative and fist positive test Risk years: 14 # seroconversions: 45 Incidence rate: 4.5 infections per 1 person years Source: Sypsa et al 217 (JID) ARISTOTLE data suggested a 78% reduction in incident HIV infections between the start and end of the program (18 months) Source: ARISTOTLE; Sypsa et al J Inf Dis., 217 5

16/6/217 Over-representation of HIV positives in OST entries coupled with reduced sample sizes in the recent years threaten data reliability of HIV prevalence in this population 6 55 5 45 4 35 3 25 2 15 1 5 All entering OST in Athens (N tests = 95 in 215) < 25 years of age (OST) (N tests = 4 in 215) <2 injection years (OST) (N tests = 1 in 215) 1st-ever-treatments (OST) (N testd = 23 in 215) 28 29 21 211 212 213 214 215 Fig. Prevalence of HIV infection among PWID entering OST in Athens Rely on HIV notifications as proxy for trends in incident HIV infection among PWID in Athens 12 1 8 HIV infections associated with injection drug use, N % of HIV infections associated with IDU (among cases with known mode) 54 8 6 Number, Ν 6 4 33 551 36 4 Percentage, % 2 266 272 18 13 13 2 4 3 2 3 3 17 11 11 15 16 16 7 58 26 27 28 29 21 211 212 213 214 215 216¹ Fig. Trends in the number and proportion of HIV notifications associated with injection drug use in Greece Source: Hellenic Centre of Disease Control & Prevention 6

16/6/217 But Fig. Trends in the number of syringes distributed and the number of syringes per estimated current PWID in Athens Syringes distributed / exchanged, N 5, 45, 4, 35, 3, 25, 2, 15, Syringes exchanged/distributed, N Syringes per estimated injection* drug user, Ν 119397 1, 64958 5519 68579 61516 5, 29782 3489 8 11 2 19 21 16 25 26 27 28 29 21 (3537) (3243) (326) (2855) (3194) (391) 46898 43 211 (283) 133 435617 212 (369) 223 31576 25 2 15 139 26761 19 1 Source: Greek Reitox Focal Point of the EMCDDA, 216 213 (1956) 214 (2272) 215 (2451) 5 Syringes per current injection drug user, N 7 * Past 3 days injection Reporting year (Estimated number of current injection drug users; central estimate) 6 551 5645 5834 5696 Number, N 4847 4786 4894 4682 474 5 4451 4269 4248 4232 487 391 363 3637 3767 4 362 374 3131 3275 358 3371 3233 316 3 2 2511 266 2135 1761 861 888 862 94 198 1122 771 1 621 535 597 59 475 442 277 376 339 367 44 47 369 328 498 254 288 247 156 Greek 22 Reitox 23Focal 24 Point 25 26 of the 27 EMCDDA 28 29 21 211 212 213 214 215 All entries (any modality) Entries in drug free Entries in OST Presentations in low threshold Fig. Trends in the number of treatment entries in Greece, total and by treatment modality Source: Greek FP ARISTOTLE enriched data on HCV prevalence by including PWID from wider networks in Athens Based on treatment entry data: Antibody prevalence has been consistently higher among those with injection careers 2 years, compared to new injectors Trend in antibody prevalence among new injectors parallels that of old, but fluctuations are sharper among new injectors, perhaps associated with their small Ns (%) 1 9 8 7 6 5 4 3 2 1 74 25 25 67 66 15 anti-hcv prevalence in PWID with <2 years injection career (Athens) anti-hcv prevalence in PWID with 2 years injection career (Athens) 63 65 63 28 28 59 41 39 62 43 71 76 54 52 78 65 5% 95% CI: 45.%, 54.7%] (N=431) 7 71 48 55 61 64 32 29 New injectors in ARISTOTLE 21 22 23 24 25 26 27 28 29 21 211 212 213 214 215 216 Fig. Trends in the prevalence of anti-hcv among PWID entering drug services in Athens, by injection careers Source: DRID data of ; Sypsa et al (poster) 217 216 data are preliminary, not to be cited 7

16/6/217 ARISTOTLE data allowed the direct measurement of also HCV incidence among new PWID in Athens Infectious disease: HCV Sample / data: Recent (past 12 months) PWID and 2 years injection history Participants of the 5-round RDS of the ARISTOTLE program N = 431 Period: 212-213 Type of measurement: Direct, longitudinal follow up of PWID who where initially negatives and seroconverted during the program Seroconversion time: Mid-point between last negative and fist positive test # risk years: 28.4 # seroconversions: 16 Incidence rate: 56.3 infections per 1 person years Source: Sypsa et al 217 (poster EASL) Focal point data considered in modelderived estimations HCV antibody prevalence (stratified) (DRID) Estimations of HRIDU population size (HRDU) Prevalence of syringe sharing among PWID (TDI / DRID) Age of onset of injecting (TDI / DRID) Duration of injection careers (TDI / DRID) Drug related deaths (DRD) NSP coverage (Responses & HRDU) OST coverage (Responses & HRDU) Length of stay in OST (Treatment data) 8

16/6/217 A possible outbreak of HCV infections among PWID in Athens from 29 through 211, followed by a decline in incident cases in 212 and stability thereafter HCV incident cases 14 12 1 8 6 4 2 22 23 24 25 26 27 28 29 Year 21 211 212 213 22,7* (17.3-27.4) Gountaset al 217 (Addiction) 214 215 Fig. Trend in the estimated number of HCV incident cases among PWID in Athens based on a dynamic disease transmission model Source: Gountas et al 217 (oral present.) Incidence 95% CRI Antibody prevalence grew together with the growth of the epidemic and after the number of the incident cases started to fall HCV incident cases, N 14 12 1 8 6 4 2 Notes Average yearly number of tests = 155 (range: 382, 1513 ) Estimated HCV incident cases based on disease transmission modelling, N anti-hcv prevalence among PWID accessing drug services, % 22 23 24 25 26 27 28 29 21 211 212 213 214 215 1 8 6 4 2 Anti-HCV prevalence, % Fig. Trends in the estimated number of HCV incident cases among PWID and antibody prevalence among PWID accessing treatment services in Athens Source: Adapted from Gountas et al 217 (oral presentation); 9

16/6/217 Higher-risk injecting behaviour has seemingly taken place prior to the HIV outbreak among PWID in Athens in 211 14 12 Newly diagnosed / reported HIV infections associated with IDU, N HCV incident cases (estimation), N 1 Number of cases 8 6 4 2 22 23 24 25 26 27 28 29 21 211 212 213 214 215 Fig. Trends in the estimated number of HCV incident cases among PWID in Athens and number of newly diagnosed / reported HIV cases associated with injection drug use in Greece Source: Adapted from Gountas et al 217 (oral presentation); Adapted from Hellenic Centre of Disease Control & Prevention Notes. Almost all HIV cases were reported in Athens Four key points 1. Strong indications for reductions in prevalence and incidence of HIV and HCV infections among PWID in Athens in the recent years 2. Positive trends likely associated with the impact of interventions and the resulting behavioural changes towards less risky injection and sexual routines 3. Infection risks are by no means eliminated in this population (e.g., ongoing economic recession, poverty, social exclusion and lack of social support, discontinuation of interventions) 4. New model-derived estimations of incidence draw on focal point data and provide new insights about the course of the epidemics among PWID in Athens 1

16/6/217 TDI / DRID data A valuable tool for measuring / estimating incident infections Strengths Annual samples Core and optional DRID and TDI variables High coverage (units and entries) Not intervention-type programs Weaknesses Biased sample (e.g, PWID interested in treatment, structural barriers in accessing services ) Small sample sizes of recurrent entries No measures about time / year of infection Missing serological data owing to delays in testing Opportunities Direct measurement of incidence: Identify PWID who record multiple presentations in treatment services over a period of 15 years. Plus, possibility to assess risk factors for seroconversion Model-derived estimations: Feed in with data on important parameters (e.g., high risk injection use, injection careers, sharing partners etc) Thank you for your attention! Cited work Gountas, I., Sypsa, V., Anagnostou, O., Martin, N., Vickerman, P., Kafetzopoulos, E., et al. (217). Treatment and primary prevention in people who inject drugs for chronic hepatitis C infection: is elimination possible in a high prevalence setting? Addiction, DOI: 1.1111/add.13764. Hatzakis, A., Sypsa, V., Paraskevis, D., Nikolopoulos, G., Tsiara, C., Micha, K., et al. (215). Design and baseline findings of a largescale rapid response to an HIV outbreak in people who inject drugs in Athens, Greece: the ARISTOTLE programme. Addiction, 11(9), 1453-1467. Sypsa V, Vickerman V, Wiessing L, Malliori M, & Hatzakis A. High levels of transmission of HCV infection among people who inject drugs in Greece (poster presented at EASL the International Liver Congress, Amsterdam 19-23 April 217) Sypsa, V., Psichogiou, M., Paraskevis, D., Nikolopoulos, G., Tsiara, C., Paraskeva, D., et al. (217). Rapid decline in HIV incidence among persons who inject drugs during a fast-track combination prevention program after an HIV outbreak in Athens. The Journal of Infectious Diseases, doi: 1.193/infdis/jix1. Indicator Drug Related Infectious Diseases, Greece Contact person: Anastasios Fotiou, PhD University Mental Health Research Institute (UMHRI) 2, Soranou tou Efesiou, st. PO Box 66 517, 156 1, Papagou Athens Greece E: afotiou@ektepn.gr T:+321653692 F:+3216537273 11