Methods and Results

During 1990–1994 and 1999, voluntary, anonymous, cross sectional surveys of Glasgow's injectors were performed to determine their prevalence of HIV, HCV and associated risk behaviours.3 In the 1990–1994 sweeps of the survey, people were eligible for participation if they had injected drugs in the two months before interview; in 1999 respondents were eligible if they had ever injected since January 1990. A multi-site, community wide, sampling strategy—with respondents drawn from both “in treatment” and “out of treatment” settings—was adopted to minimise selection bias. The questionnaire administered by trained interviewers was standardised across survey years. An assessment was made of changes in needle/syringe sharing over successive cross sectional surveys, using multifactorial logistic regression analyses. Data were analysed using S-PLUS software.

Questionnaires were obtained from 2966 individual injectors (duplicate interviews within and between years were removed before analysis). The sample size was approximately 500 in each year between 1990 and 1994 (table 1); 444 respondents were recruited in 1999 but 54 people were eliminated from the analysis because they had not injected in the two months before interview and were therefore not comparable with the previous samples. Seventy one per cent of the total sample were male and 77% injected at least twice daily; their mean age was 26 years. In the multifactorial regression analyses, after adjusting for behavioural and demographic characteristics (including time since first injection to allow for the shorter injecting careers of those recruited in 1999), a significant decline was observed between years 1990 and 1992, in the risk of reporting injecting with used needles/syringes (a) at least once, (b) at least weekly, (c) with more than one person and (d) with a casual acquaintance, in the previous six months (table 1). There was also a significantly increased risk of reporting behaviours (a–d) between 1992 and 1999 (table 1).

Discussion

Throughout the 1990s the Greater Glasgow Health Board developed the city's needle/syringe exchange services for its estimated 6000–8500 injectors4; the numbers of needles/syringes distributed to injectors increased from 136 900 in 1990 to 238 500 in 1992, 238 926 in 1994, 383 870 in 1998 and 787 524 in 1999. Despite this increase in the provision of injecting equipment, particularly in the late 1990s, needle/syringe sharing—as measured by several parameters and adjusting for behavioural and demographic characteristics, including time since first injection—increased significantly between 1991 and 1999. These observations confirm other less robust indicators of increased sharing activity and are cause for considerable concern. Conditions for the frequent transmission of HCV and HBV, and a resurgence of HIV, among injectors are extremely favourable. The reasons why such risks are being taken by a generation of injectors that are different from that of the mid-1980s, need to be elucidated and tackled accordingly. The provision of sterile injecting equipment is essential in preventing the spread of bloodborne infections; however, unless injectors also fully appreciate the dangers of needles/syringe sharing, their preventive effectiveness will be severely attenuated. The Italian experience of the early 1980s5—when epidemic spread of HIV, HCV and HBV affected tens of thousands of injectors who had no difficulties in accessing sterile needles and syringes—is a salutary warning.