Every year approximately 30 million people are sentenced to imprisonment worldwide. Whilst most will return to their communities, the period after release is a high risk for various health outcomes as well as mortality. In the first two weeks of prison release, the risk of death of former inmates is almost 13 times more likely than that of the general population (Binswanger et al, 2007).

Psychiatric disorders and substance use disorders are highly prevalent in prison systems. A 2006 systematic review demonstrated prevalence rates between 17.7 to 30.0% in males and between 10.0 to 23.9% in females for alcohol abuse/dependence (Fazel et al, 2006). Despite this, prison represents an environment in which these disorders may be potentially treated in a marginalised population, who are out of reach of conventional healthcare systems.

Little is known about the long term effects of these disorders following release from prison. In a recent longitudinal study published in The Lancet Psychiatry, the association between psychiatric disorders and post-release mortality was investigated (Chang et al, 2015).

Methods

Data from population-based registers in Sweden were used and linked with personal identification numbers. The authors selected a cohort of individuals who were imprisoned since January 1^st 2000, and released prior to December 31^st 2009. Individuals were followed up until death, emigration or the end of the study.

The main outcome measured was death after release from prison, which was identified using the Swedish Cause of Death Register, and the underlying and contributing causes of death were coded. Covariates such as criminological factors (e.g. length of incarceration), sociodemographic factors (e.g. employment, education) and familial factors were also measured.

The population attributable fraction (PAF) was used to assess the effect of psychiatric disorders on port-release mortality. The PAF measures the proportion of deaths attributable to a risk factor, assuming causality.

Results

The authors identified 47,326 individuals (43,840 men) in prison during the study period and followed up for 238,457 person years after release. 2, 703 (6%) men and 171 (5%) women died after release from prison with the median time from release to death at 3.5 (1.7 to 5.4) years. Overall all-cause mortality was 1,205 deaths per 100,000 person years.

Substance use disorders significantly increased the rate of all-cause mortality

• Alcohol use: adjusted HR 1·62, 95% CI 1·48 to 1·77
• Drug use: adjusted HR 1·67,95% CI 1·53 to 1·83

Substance use disorders also increased external-cause mortality:

• Alcohol use: adjusted HR 1·54, 95% CI 1·34 to 1.78
• Drug use: adjusted HR 2·43, 95% CI 2·43 to 2.78

Furthermore these associations remained after controlling for sociodemographic, criminological and familial factors.

Assuming causality, the following numbers of death were potentially attributable to substance use:

• 925 out of 2,703 deaths in males (PAF = 34%)
• 85 out of 171 deaths in females (PAF = 42%)

Associations between other psychiatric disorders and all-cause mortality were significant when adjusting for age, sex and immigration status. However, they disappeared when substance use disorders were controlled for.

Discussion

Individuals who are released from prison show substantially increased rates of mortality if previously diagnosed with a substance use disorder. Importantly, these associations remained when controlling for sociodemographic, criminological and familial factors, adding support for a causal relationship. The study found little evidence that other psychiatric disorders such as Attention Deficit Hyperactivity Disorder or depression/anxiety increased post-release mortality.

The authors speculate that the increased risk of mortality due to substance use disorders may be attributable to the diminished physiological tolerance to substances after release. This in turn may lead increase the risk of overdose. Indeed, previous research suggests overdose is the leading course of death among previous inmates, contributing to a quarter of all deaths. The authors suggest that future research is needed to examine cost-effectiveness of substance use treatment in prison settings, but also to identify best service models to maintain treatment upon release.

The study has several important implications for public health. For example, specialist services for addiction in prisons have the potential to make a considerable impact on mortality whether through direct treatment whist in custody or encouraging contact with community services upon release. Furthermore, these findings may help identify ‘high risk’ groups of individuals upon entering the prison system.

Limitations

• The analyses did not account for the effects of any treatment of substance use during incarceration, which may actually have led to conservative estimations of mortality.
• Secondly, care is needed when generalising these findings to other countries, which may have different criminal justice systems. For example, Sweden has much lower incarceration rates compared to other countries despite similar prevalence of substance use disorders (Sweden’s incarceration rate is approximately half of the UKs).

Conclusion

To conclude, this study demonstrates alcohol and substance use disorders may increase the risk of mortality upon release from prison. Reductions in mortality may be seen if effective treatments are administered during incarceration and there is contact with community support upon release.

Links

Primary paper

Chang, Z., Lichtenstein, P., Larsson, H., Fazel, S. (2015). Substance use disorders, psychiatric disorders, and mortality after release from prison: a nationwide longitudinal cohort study. The Lancet Psychiatry, 2(5): 422-430.

Other references

Fazel, S., Bains, P., Doll, H. (2006). Substance abuse and dependence in prisoners: a systematic review. Addiction, 101(2): 181-189.

Binswanger, I., Stern, M., Deyo, A., et al (2007). Release from Prison – A High Risk of Death for Former Inmates. New England Journal of Medicine, 356(2): 157-165.

Photo Credit: dcosand via Compfight cc