Table 2

Trends in injecting and non-injecting drug use behaviours, using 13 620 visits of 996 drug users in the Amsterdam Cohort Study

86–88 89–90 91–92 93–94 95–96 97–98 Yr Yr2-150Yr
Injecting status (%)
 never111113141821
 stopped222628313542
 current666459554736NS 2-160
Transitions injecting: (%)2-151
 initiation4.15.12.81.11.10.7 2-150 NS
 cessation10.010.210.111.013.317.1NS 2-165
 relapse21.320.119.316.913.411.8 2-165 NS
Injecting behaviours: (%)2-152
 Main type:heroin202120172224
cocaine91210152018 2-160 NS
speedballs615660595145
other1011109713
 ⩾2 injections/day393632312522 2-165 NS
 All needles at exchange576566757373 2-165 2-165
Non-injecting use: (%)
 heroin:<daily343233364342 2-165 2-165
daily271915182019
 cocaine:<daily302819313848 2-165 2-165
daily1410981014
 tranquilisers442833323543 2-165 2-165
 >4 units alcohol/day221819232120NSNS
  • Yr, p value effect calendar year; Yr2-150Yr, quadratic effect calendar year;

  • 2-150 , p<0.05;

  • 2-160 , p<0.01;

  • 2-165 , p<0.001; GEE was used for statistical analysis, using a linear link function.

  • 2-151 Expressed as percentage per visit at risk, see methods.

  • 2-152 Only at visits at which participants reported current injecting.