BACKGROUND: We report on an exercise to estimate the prevalence of injecting drug use (IDU) and associated harms in a single primary care trust. METHODS: Covariate capture-recapture methods to estimate (i) IDU prevalence; respondent driven sampling to measure (ii) prevalence of HCV and HIV and record linkage to measure (iii) mortality risk. RESULTS: (i) The overall estimated number of IDU was 5540 (95% confidence interval, CI: 4710-6780) for all cases and 3280 (95% CI: 1940-4610) for cases matched to primary care register, i.e. a prevalence of 2.2 and 1.3% aged 15-54, respectively. (ii) The prevalence of HCV, hepatitis B and HIV was: 53, 32 and 0.7%. Over 70% of IDU in Bristol reported having at least one vaccination for HBV; more than half of those who were HCV positive were undiagnosed. (iii) The all-cause and overdose mortality rates for IDU were 0.75 and 0.4% respectively; and the standardized mortality ratio was 7.8 (95% CI: 5.4-10.8). CONCLUSION: Locally specific and useful intelligence on injecting and its health consequence can be generated to inform local public health action, and may contribute information to validate national prevalence estimates.
BACKGROUND: We report on an exercise to estimate the prevalence of injecting drug use (IDU) and associated harms in a single primary care trust. METHODS: Covariate capture-recapture methods to estimate (i) IDU prevalence; respondent driven sampling to measure (ii) prevalence of HCV and HIV and record linkage to measure (iii) mortality risk. RESULTS: (i) The overall estimated number of IDU was 5540 (95% confidence interval, CI: 4710-6780) for all cases and 3280 (95% CI: 1940-4610) for cases matched to primary care register, i.e. a prevalence of 2.2 and 1.3% aged 15-54, respectively. (ii) The prevalence of HCV, hepatitis B and HIV was: 53, 32 and 0.7%. Over 70% of IDU in Bristol reported having at least one vaccination for HBV; more than half of those who were HCV positive were undiagnosed. (iii) The all-cause and overdose mortality rates for IDU were 0.75 and 0.4% respectively; and the standardized mortality ratio was 7.8 (95% CI: 5.4-10.8). CONCLUSION: Locally specific and useful intelligence on injecting and its health consequence can be generated to inform local public health action, and may contribute information to validate national prevalence estimates.
Authors: Beat Müllhaupt; Philip Bruggmann; Florian Bihl; Sarah Blach; Daniel Lavanchy; Homie Razavi; David Semela; Francesco Negro Journal: PLoS One Date: 2015-06-24 Impact factor: 3.240