Impact of scaling up harm reduction interventions on injecting risk behaviours, ART outcomes and HIV incidence among people who inject drugs in Kenya.

INTRODUCTION

Little data exists on the effectiveness of HIV prevention interventions among people who inject drugs (PWID) in Africa. We used empirical data from Kenya to fill this evidence gap.

METHODS

Six rounds of bio-behavioural surveys using respondent-driven-sampling were conducted among PWID in Nairobi and Coastal Kenya over 2012-2015. Dried blood spot samples were tested for HIV and HIV viral load, and HIV incidence was estimated through linking participants between rounds. Regression analyses evaluated whether self-reported usage of opioid agonist therapy (OAT) or needle and syringe programmes (NSP) in last year were associated with reduced injecting risk behaviours, increased ART uptake and viral suppression, and reduced risk of HIV acquisition.

RESULTS

Overall, 4897 PWID participated in the study, with 3903 participating in >1 round. Over the rounds, coverage increased from zero to 80-86 % for NSP and zero to 10-20 % for OAT. The proportion of people living with HIV (PLHIV) that were virally suppressed increased from 7-14 % to 39-55 %. Accessing NSP and OAT was associated with reduced syringe sharing at last injection (NSP adjusted odds ratio (aOR)=0.31; 95 %CI:0.24-0.40; OAT aOR=0.046; 95 %CI:0.034-0.061) and OAT was associated with reduced injecting frequency (adjusted rate ratio=0.21; 95 %CI:0.12-0.36). Using OAT was associated with increased ART coverage (aOR=2.76; 95 %CI:1.50-5.06) and viral suppression (aOR=2.99; 95 %CI:1.78-5.03) among PLHIV, while NSP was not. HIV incidence decreased from 6.10 (95 %CI:3.56-9.77) to 1.49 (95 %CI:0.79-2.54) per 100 person-years between the first and second half of the study. Accessing NSP was associated with lower HIV incidence (adjusted hazard ratio=0.25; 95 %CI:0.087-0.58).

CONCLUSIONS

This study provides strong evidence for the benefits of NSP and OAT on varied HIV outcomes among PWID in Africa.