Substance misuse is an acknowledged public health problem which continues to grow as a worldwide issue.1 Substance misuse is defined as ‘recurrent use that is causing actual harms (negative consequences) to the person (including dependence, but also other health, psychological or social problems), or is placing the person at a high probability/risk of suffering such harms. As such, substance misuse contributes to a range of adverse health outcomes including other diseases (e.g. infectious diseases) and increased risk of suicide, violence and injury. All of which can directly or indirectly increase risk of mortality.2 Substance misuse is a leading cause of premature mortality and morbidity.3 This Data Insight presents the findings of a study conducted as part of the BOLD Substance Misuse Demonstrator Pilot (Phase 1) in Wales, which focuses on early intervention and prevention of escalation of substance misuse. Further information on the BOLD programme can be found here: Better Outcomes Through Linked Data (BOLD). This research, focussing on mortality, forms part of a larger study examining the type and timing of healthcare service contact prior to the start of problematic or enduring substance misuse in order to identify missed opportunities for prevention of escalation and promote earlier engagement to reduce substance misuse related harm including death.

What we did: We used a retrospective population-based cohort study design to identify a Welshresident cohort with evidence of substance misuse (aged ≥10 years) who died between 2010 and 2019 by linking different health datasets. We used routinely collected administrative health data from the Secure Anonymised Information Linkage (SAIL) Databank for the study. Datasets used include primary care, hospital admissions, emergency department (ED), specialist substance misuse treatment and death records. In this study, we analysed visits to health services related to substance misuse across different types of healthcare service providers in Wales. We examined how these visits affected the subsequent risk of all-cause death, death due to drug misuse, death from alcohol-specific causes, and suicide. An individual was defined as having a substance misuse event if their GP or hospital records contained a code relating to alcohol and/or drug misuse. Drug misuse was further stratified into: ‘opioids’, ‘stimulants’, ‘cannabinoids’ or ‘other’. The Office for National Statistics (ONS) code lists were used to identify alcohol-specific and drug misuse deaths. Suicide was established by using previously validated codes developed by the Adolescent Mental Health Data Platform. We analysed recorded substance misuse events within each of the health datasets separately to investigate whether the point of contact affected mortality risk. For example, an individual first seen in ED, and later during a hospital admission (secondary care) would be included in both ED and secondary care cohorts. We estimated the risk of mortality (all-cause, alcohol-specific, drug misuse death and/or suicide) amongst individuals with a history of substance misuse. We also assessed whether the timing of the first health contact related to alcohol and/or drug misuse affected mortality. We adjusted the models for sex, age at first substance misuse, ethnicity and Welsh Index of Multiple Deprivation (WIMD) quintile. 

What we found: A higher proportion of individuals with a substance misuse-related health event died prematurely compared to the general population. Substance misuse is associated with increased risk of premature death from all causes, potentially due to the impact of substance use on physical and mental health more broadly, and deaths that are specifically substance misuse related and suicide. During the study period, 16,229 people died. A high proportion of these individuals died prematurely, aged under 75, compared to average life expectancy rates1. For males and females in the cohort, all-cause mortality rates in those aged under 75 years were 69.3% and 61.5% respectively, compared to rates in the general population of 39.8% (males) and 26.4% (females) over the same period.2 Cardiovascular and respiratory diseases were the most common underlying causes of death. However, within the top ten underlying causes of death, two were directly related to alcohol consumption: alcoholic liver disease (3.1%) and alcoholic hepatic failure (2.6%)....