Twenty-one primary studies were included in the review, most of which were cross-sectional or interrupted time series. Of the 120 potentially relevant reports identified by the search strategy, 36 were unavailable. Six of the 84 available reports were included without a formal assessment of relevance, as they were invited commentaries and letters to the editor relating to one key primary study, and so were included as addenda to the key study. Twenty-five of the remaining 78 studies were assessed as relevant: 21 primary studies and 4 'reviews' (a review, an editorial, an update and a position paper). All 4 reviews were assigned a 'weak' quality score and were subsequently excluded from the review.
One primary study was assigned a quality rating of 'moderate', while the remaining 20 were assigned a quality rating of 'weak'.
Behavioural outcomes.
Seven of the 9 studies investigating the prevalence of needle and syringe sharing practices among IDUs found that NEPs were effective at reducing the number of IDUs sharing needles and syringes. One study found a non-significant difference in needle sharing between frequent and non-frequent attenders. However, there was a significant reduction in the number of women IDUs sharing needles and syringes. One study found that, among hepatitis B and hepatitis C susceptible IDUs, a significantly higher proportion of regular NEP users reported sharing syringes than IDUs who had never used the NEP.
Three of the 5 studies investigating the prevalence of sharing drug preparation equipment found that NEPs were effective at reducing the number of IDUs sharing drug preparation equipment.
Two studies investigating the number of times a needle was used before being discarded. Both found that the number of times a needle was used decreased among NEP users. One study investigated the practice of passing on used needles or syringes to another user and found that the NEP was effective at reducing this practice.
Two studies did not find an increase in the use of bleach to clean needles following attendance at an NEP.
Two of the 3 studies investigating the frequency of injection found that the frequency reduced in NEP users; the other study reported a significant increase in injection frequency.
Physical status outcomes.
Only one of the 3 studies reporting HIV seroprevalence reported a decline in NEP attendees. One study reported higher rates of HIV infection in frequent attenders at NEPs than in non-frequent attenders. The other study reported that baseline HIV seroprevalence was significantly higher at enrolment for attenders than non-NEP attenders. Two studies indicated that NEP participation has no apparent protective effect against the hepatitis B virus.
Programme process outcomes
Three out of 5 studies found improvements in the number of sterile syringes available to IDUs. In contrast, 2 studies suggested that NEPs were not meeting the needs of IDUs. Two studies found improvements in terms of the number of used syringes in circulation.