Search

Objective To identify causes of death among Norwegian drug abusers and to investigate the risk factors for fatal overdose and other causes of death, with specific attention to ageing and duration of abuse. Methods In a cohort of 501 drug abusers admitted to treatment in the period 1981–1991, mortality has been calculated as incidence rates. The analyses of time to death were conducted as proportional hazard regression models using a competing risk approach. Results Crude incidence rates for all deaths and overdose deaths did not vary with age. For non-overdose deaths, however, the incidence was significantly higher after the age of 40. Explanatory factors associated with age at fatal overdoses are also associated with age at death by other causes. At every age the risk of death was higher with a long-term abuse of drugs, and more so for fatal overdose than for death by other causes. Conclusions With respect to fatal overdose duration of abuse, but not ageing, is found to be a risk factor. With respect to death by other causes both ageing and duration of abuse are factors associated with such death.

A rough estimate indicates that women amount to one third of the heroin user population in Norway. However, among the youngest users females account and have accounted for a larger proportion. Given this greater representation of females in the younger age groups one would expect that the gender difference in the general population of users would be reduced over time due to a cohort effect. However, the ratio of male to female users has remained stable. The prevailing explanation for this stability is based on the claim that women are more likely to overcome their abuse. This paper addresses both the basis for the assumption of a missing cohort effect and examines an alternative explanation linked to the way into abuse rather than the way out. The study primarily draws upon two distinct data sources. The first body of material consists of interviews carried out in the period 1993–1997 with people attending the only needle exchange service in Oslo. The second data source consists of police records, which provide the number of persons arrested each year in the period 1984–1997 for using, possessing or selling heroin distributed by age and gender. We find that the age of injection debut has increased and that this is especially true for male users. Applying a simple quantitative model, the effect a change in recruitment may have on the male-to-female ratio is analysed. The results indicate that a higher debut age may have contributed to stabilizing the gender gap. The finding suggests that more research is needed on the conditions of recruitment in order to better predict the future population demographics and to avoid hasty and incorrect conclusions.

The purpose of this study was to explore to what extent alcohol is a factor in non-fatal overdoses on the basis of records of ambulance emergencies, and to what extent this varies across gender and age. Furthermore, we wanted to investigate whether alcohol intake, in relation to an overdose, is associated with the risk of recurrent overdoses, and if so, whether such an association varies across gender and age. To investigate the role of alcohol intake in non-fatal overdoses, analyses were conducted both at the event level and at the individual level. Bivariate associations were explored in table analyses and by comparisons of means. To determine whether alcohol intake was associated with experiencing recurrent overdoses, survival analyses were conducted applying Kaplan-Meier estimation and Cox regression models. Furthermore, a linear regression model was estimated to assess the impact of gender age and number of overdoses on the proportion of overdoses in which alcohol was involved. Between groups of clients who had overdosed once, several times or many times, we found that there was a U-shaped relationship. The proportion of overdoses with alcohol involved was highest among those who had overdosed once and those who had overdosed more than ten times. The probability for a recurrent overdose was higher among those who were reported with no alcohol intake in the first overdose. Being female and having alcohol involved in the first overdose registered during our observation period reduced the risk for a recurrent overdose. However, age seemed to be a gradient with respect to alcohol’s association with recurrent overdoses. While alcohol was associated with a significantly lower risk for recurrence in the two youngest age groups, this is not the case in the oldest age group. A possible explanation might be that it is a change in the pattern of drug use as an effect of aging where infrequent heroin use in combination with frequent alcohol intake increases with increasing age. For this type of drug users the individual’s risk of recurrent overdoses may be lower due to fewer events of heroin intake.