Study funded by NIHR Health Technology Assessment ref: 13/17/04 Improving injecting skills and preventing blood borne virus infection in people who inject drugs in the UK: A feasibility randomised control trial of a psychosocial intervention (PROTECT) Gail Gilchrist Gail.Gilchrist@kcl.ac.uk Study funded by NIHR Health Technology Assessment ref: 13/17/04

Declarations No conflicts of interest Study funded by NIHR Health Technology Assessment ref: 13/17/04

The Research Team King’s College London: Gail Gilchrist, Davina Swan, John Strang University of the West of Scotland: April Shaw, Alison Munro, Avril Taylor University of York: Ada Keding, Steve Parrott, Judith Watson Betsi Cadwaladr University Hospital Trust: Sarah Towers Public Health Wales: Noel Craine University of Huddersfield: Elizabeth Hughes

Recent outbreak of HIV among PWID in the UK Background In the UK, 23%-61% of people who inject drugs (PWID) are hepatitis C (HCV) positive, the rates of HIV and hepatitis B (HBV) are much lower, 0%-1.4% for HIV and 6-18% for HBV (Hahné et al., 2013; PHE, 2015) Recent outbreak of HIV among PWID in the UK Increased risk of infection for those who inject amphetamines and amphetamine-type drugs, such as, mephedrone (Hope et al., 2016) 4

Background Opiate substitution therapy and needle exchanges have reduced HIV and HCV among PWID (Degenhardt et al., 2010; Turner et al., 2011; Aspinall et al., 2014; MacArthur et al., 2014) Public Health England’s, “Shooting Up” report highlighted that in 2014 in England, Wales and Northern Ireland, sharing of needles in the previous month was reported by 17% of individuals attending drug treatment services, in Scotland this figure was 14% in 2013-2014. Addressing risks and increasing knowledge to reduce BBV among PWID remains a public health priority Alongside harm reduction provision, Psychosocial interventions (such as brief interventions, motivational interviewing, cognitive behavioural therapy and contingency management) could further decrease BBVs (Degenhardt et al., 2010) by educating PWID about transmission risks and developing strategies to avoid them. 5

Aims of the PROTECT study to understand current influences on PWID BBV risk behaviours to develop an evidence-based psychosocial intervention aimed at reducing BBV transmission risk behaviours and increasing BBV transmission knowledge among PWID to explore the feasibility of recruiting PWID to a trial comparing the intervention to control to explore the feasibility and acceptability of delivering a psychosocial intervention to PWID 6

Aims of the PROTECT study to understand current influences on PWID BBV risk behaviours to develop an evidence-based psychosocial intervention aimed at reducing BBV transmission risk behaviours and increasing BBV transmission knowledge among PWID to explore the feasibility of recruiting PWID to a trial comparing the intervention to control to explore the feasibility and acceptability of delivering a psychosocial intervention to PWID 7

The PROTECT Intervention (evidence based and informed)

How the intervention was developed Building on evidence of “what works” – review and adaptation of existing interventions [Gilchrist et al., 2017 AIDS & Behavior] Focus on what is important to PWID from our indepth interviews: Interplay of structural, situational and individual factors influenced injecting risk behaviours Focus for PWID not on BBV rather on avoiding abscesses, finding a vein etc Interventions should target other injecting-related priorities including improving injecting techniques and venous care to promote the use of sterile injecting equipment, and protective strategies to avoid risk 9

How the intervention was developed Current thinking – addressing “symbiotic ” goals for PWID, such as avoiding injecting related scars or marks and maintaining venous access, may result in the use of sterile injecting equipment (Harris & Rhodes, 2012) Inclusion of protective practices and strategies to avoid injecting risk situations such as withdrawal and lack of preparedness (Harris et al., 2012; McGowan et al., 2013; Mateu-Gelabert et al., 2014; Treolar et al., 2015).   Co-development with experts (service providers, service users, peer educators, researchers, policy makers) 10

The PROTECT intervention (3 x 1 hr sessions) Improving injecting techniques, good vein care 2 Session 2 Planning for risk situations 3 Session 3 Understanding BBV transmission risks Sessions used videos, games and exercises to facilitate discussion and build skills and strategies to reduce and avoid risk. All sessions also included a short didactic education section. Separate groups were held for women and men.

Download the PROTECT intervention https://www. kcl. ac Download the PROTECT intervention https://www.kcl.ac.uk/ioppn/depts/addictions/research/drugs/PROTECT-download-page-form.aspx or email Gail.Gilchrist@kcl.ac.uk

The feasibility trial

Recruitment Researcher recruited; posters/flyers; staff facilitated recruitment in London, York, Glasgow and North Wales PWID (not just PEDs) in past month, ≥ 18 years from community substance use and harm reduction services, and needle exchange programmes; able to speak English 14

Assessment Demographic data Injecting and sexual risk behaviours and self-efficacy Withdrawal Prevention Tactics scale tactics to avoid withdrawal episodes: saved a bag for the next morning; put aside additional drugs; stored methadone; or put aside money for getting the next bag in an emergency (Vazan et al., 2012). A fifth item asked about use of other substances, such as painkillers, to avoid withdrawal symptoms until they are able to obtain their drug of choice. BBV transmission knowledge Motivation to change behaviour Health-related Quality of Life (HRQoL) (Rabin & Charro, 2001) Health and social resource used

Intervention groups trial After baseline assessment, randomised to: Treatment group PROTECT 3 x 1 hr sessions of psychosocial intervention plus Hep C Info booklet & leaflet about HIV outbreak Control group Hep C Info booklet & leaflet about HIV outbreak only Both groups also received treatment as usual 16

Contingency management £10 for attendance at each of the 3 PROTECT intervention sessions + “Bonus” £10 if attended all 3 sessions Participants in each treatment arm received £10 for research interview/ and participants in treatment arm received £10 for participating in focus group Cash (London) versus Voucher (all other areas)

Settings and facilitators London Co-facilitated by drugs worker & peer educator (gender matched) Recruited from 2 drug treatment services/ NEX, and one homeless hostel with prescribing clinic Delivered in one drug treatment service/ travel reimbursed York Groups due to be facilitated by male BBV nurse for male group and female drug worker Due to be delivered in a drug treatment service 18

Settings and facilitators Glasgow Men’s group co-facilitated by 1 female service coordinator/1 male group worker (no female group took place) Recruited from 1 drug treatment services/ NEX Delivered in one drug treatment service North Wales Groups co-facilitated by 1 x male & 1 x female harm reduction worker Recruited from drug treatment service, homeless drop in centre and needle exchange & mobile harm reduction van Delivered at homeless drop-in/ travel reimbursed 19

Baseline characteristics   Male Female Intervention N=34 Control N=30 N=18 N=17 Age Mean (SD) 41.7 (6.81) 41.4 (7.30) 35.8 (6.06) 37.9 (8.79) Median 42.5 42 35 37 Min, Max 26, 57 22, 54 26, 48 26, 62 Number of years injecting 21.4 (8.00) 19.5 (9.01) 11.9 (7.58) 16.1 (12.14) 22 19 11.5 14 3, 36 1, 42 0, 34 0, 44 Used needle exchange in the last month 31 (91.2%) 26 (86.7%) 16 (88.9%) 16 (94.1%) Homeless 15 (44.1%) 14 (46.7%) 8 (44.4%) 5 (29.4%) HIV Positive 1 (2.9%) 0 (0.0%) Hepatitis C Positive 17 (50%) 15 (50%) 5 (27.8%) 6 (35.3%)

Feasibility 56% (99/176) of eligible participants were randomised into the feasibility trial during January and February 2016; 52 randomised to intervention and 47 to control group [34 men and 18 women] 38% (20/52) PWID attended at least one session [18 attended one session, 15 attended two sessions and 15 attended three sessions] Men were more likely to attend at least one intervention session than women (44% versus 28%). Attendance to at least one intervention session was highest in London (63%) and North Wales (54%), whereas only 25% attended in Glasgow, and no participants attended in York.

Results Those who did not attend any sessions (n=32) were more likely to be homeless (56% vs 25%, p=0.044), have injected on a greater number of days in the last month (median 25 vs 6.5, p=0.019) and used a greater number of needles from a Needle Exchange in the last month (median 31 vs 20, p=0.056). Glasgow and York had higher levels of homelessness (68% and 52% respectively) compared to London (27%) and North Wales (29%). In addition, participants injected for a greater number of days and used more needles from a Needle Exchange

Follow-up 42% (22/52) of PWID in intervention group and 49% (23/47) of PWID in control group were followed-up one month post intervention Follow-up was also highest in London (83%) and North Wales (63%), and significantly lower in Glasgow (55%) and York (43%) Follow-up was associated with fewer days of injecting in the last month (median 14 vs 27, p=0.030) and fewer injections of cocaine (13% vs 30%, p=0.063)

Feasibility trial outcomes Improved (fewer) injecting risk practices, improved self-efficacy, better hepatitis C and hepatitis B transmission knowledge and greater use of withdrawal prevention techniques in the intervention arm. At one month post-intervention amongst participants who had attended at least one session of the intervention: no increase in self-reported injecting in more “risky” sites (e.g. groin, neck) trend towards injecting on fewer days in the past 28 days Intervention does not appear to encourage riskier injecting practices or increase frequency of injecting

Acceptability to staff and participants Separate focus groups with facilitators and participants Evaluation sheets following each session completed by facilitators and participants

Possible explanations for site differences participants in some areas more complex needs (homeless, more frequent injectors etc) peer-educators co-facilitated the intervention in the London site only reimbursement for travel costs (London and North Wales), time and contingency management were paid in cash in the London site versus high street vouchers at the other three sites

Implications Complex needs of many PWID may have limited engagement of those potentially most at risk of engaging in BBV transmission behaviours women, homeless, more frequent injectors least likely to attend some participants were in prison, hospital or residential rehabilitation and therefore did not attend sessions and were not followed-up Alternative intervention delivery modes/ local solutions/ flexibility in delivery required to ensure greater reach Need for a greater embedding of BBV risk reduction in the work of substance use services/ needle exchanges

Conclusions Intervention was acceptable to both facilitators and attending participants and 57% of eligible participants agreed to be randomised, suggesting support for addressing BBV risk behaviours among PWID Considerable difficulties recruiting particular groups of PWID, mainly women and new injectors. additional barriers for women secondary distribution for new injectors PROTECT intervention has potential to positively influence some PWID BBV risk behaviour, non-attendance at the York site substantially influenced the results

Download the PROTECT intervention https://www. kcl. ac Download the PROTECT intervention https://www.kcl.ac.uk/ioppn/depts/addictions/research/drugs/PROTECT-download-page-form.aspx or email Gail.Gilchrist@kcl.ac.uk