A call to debate the decriminalisation of cannabis in the States of Jersey (the parliament of the self-governing British Crown Dependency 15 miles off the French coast) was unlikely to make the headlines around the world.¹ It does, however, represent in microcosm the debates and uncertainties that are occupying the legislators and public health authorities in many of the world’s developed insurance markets.

Canada was the first major industrialised country to allow legal, regulated access to cannabis for non-medical purposes. Since the 2017 legalisation cannabis use has grown.² The legislation came with health warnings about risks to many aspects of health, with particular concerns about daily or near daily use.³ A March 2024 review was broadly supportive of the legislation but called for increasing attention to public health awareness, focusing particularly on adolescents and young adults who would be most likely to experience harm.⁴ Germany has followed suit and the U.S. is taking “monumental” steps towards the reclassification of cannabis to a schedule III drug.^5,6

The use of cannabis as a medical treatment continues to grow, in terms of both the volume of medication and the range of illnesses in which it is employed. The evidence base for use in many conditions was described as “uncertain at best”. Neuropathic pain, chemotherapy induced vomiting and spasticity in multiple sclerosis are identified as having the best evidence of effectiveness.⁷ The same commentator raised the question of the potential for long-term harm with medicinal use, and concluded that this remains uncertain.

Against this turbulent background it is clear that attitudes to cannabis are changing and that insurers’ thinking on its use needs to reflect the changes in society.

The problems of recreational drug use and the inappropriate use of prescription medications are well known and a significant cause of concern to the insurance community. There has always been an understanding that misuse of some substances can have a significant impact upon mortality (see table),⁸ but the data upon which this work is based often come from specialist services or databases that are far from representative of an insured population.

Insurers’ response to current use of amphetamines, cocaine and non-medical opioids is clear and based on the very clear evidence of the levels of risks involved in their consumption.^9,10

The bald figures in Table 1 give a sense of the significance of the problem, but simple translation to ratings at underwriting for any given applicant reporting cannabis use, in all its forms, is problematic.

Most commonly we see people who used substances recreationally during their early years but who, with maturity, no longer indulge. To contrast the “ever-used” against the “used in the last year” numbers: it’s approximately 120 million vs 37 million, respectively, in the U.S.^11,12

This situation challenges the underwriter to identify the category of risk presented. Those who have low-level recreational use, and are at low risk of future harm; those who continue to use the drug in a way that carries the risk of long-term damage (harmful use); and, less commonly, those who demonstrate hazardous use, where there is current, obvious damage to health and social circumstances. The fourth important group are those who experience substance use disorder with compulsion, loss of control, tolerance as defined in DSM‑5 TR or ICD‑11.^13,14

The relative contribution of drug misuse to all deaths and years of healthy life lost remains relatively small globally, despite an increase of 23% in use over 10 years. Drug use accounts for 5% of all substance-related deaths and 9% of disability-adjusted life years. In terms of absolute numbers, tobacco has a far greater impact, with alcohol use a fair way behind (despite the relative risks) but it is still more impactful than other drugs.^15,16 At an individual level and in particular markets, however, the impact is much more acute and deeply troubling.¹⁷

Cannabis does not feature as a significant cause of death in the most recent EU report on drug-related deaths, despite its being “by far the most commonly consumed illicit drug in Europe”. It is estimated that 8% of EU adults used cannabis in the last year and about 1.3% (3.7 million) use the drug daily or almost daily, largely in 15 to 34‑year-old age group.¹⁸

The situation has become more complicated with changes in legislation concerning cannabis use and its increasing role in the management of medical conditions. In the American states where cannabis has been legalised for medicinal purposes there has been a 4.5‑fold increase in registrations for use. Chronic pain is the most common qualifying condition (60.6%).

In some states where recreational use has also been legalised the demand actually fell.¹⁹ Cannabis use has been shown to reduce the use of prescription of opioids and improve the quality of life for many “authorised patients”.²⁰ Small increases in the total numbers of people with cannabis use disorder have been identified in those with chronic pain where medical use was legalised – a 0.135% increase. Where recreational use was legalised, the total increased by 0.188%, with the largest increase in older adults aged 65 to 75.²¹

Despite these positive points, careful review of all pain relief at underwriting is advised, with particular attention to opioid use.

Concerns that cannabis is a “gateway” drug are longstanding and often coupled with the assumption that users of cannabis are likely to be using other more harmful substances. This concept had its clearest formulation in Kandel’s 1975 paper “Stages in adolescent involvement in drug abuse” but it probably has a much longer history.²²

There is evidence that use of cannabis is associated with the abuse of other drugs, but this needs to be put in perspective. Of those who used cannabis in the U.S., only 1.2% abused the substance, with a much smaller number abusing other drugs. Overall, 98% of cannabis users did not abuse either cannabis or other drugs.²³

Despite the small numbers involved, it is clear, that cannabis use frequently predates opioid use. In many countries non-medical opioid use is rarely reported without prior cannabis use, but this pattern is not set in stone. In Japan and Nigeria, where cannabis use is relatively rare, opioids are more frequently used without prior substance abuse.²⁴

The research about the “gateway” is of variable quality, with methodological problems (self-report, cross-sectional studies, biased sampling) and confounding variables that make conclusions difficult to draw. Confounding variables included mental health issues, socioeconomic status, and peer drug use. The Federal Research Division within the United States Library of Congress concluded that no causal link between cannabis use and the use of other illicit drugs can be claimed at this time.²⁵

The lay reporting of cannabis studies can also suggest causal links to harm when none can be demonstrated. A paper that was reported under the headline “Study Says Casual Pot Use Harmful to Teens”, strongly suggested causal links where none was demonstrated.^26,27 The possibility that disadvantaged and distressed teenagers use more cannabis than others as a potential explanation was not clearly explored.

This is not to dismiss the potential harms of cannabis use, especially in young people while their brains are developing, but to note that it is important to ensure that the medical information is properly understood, and potential risks for the applicant are fairly assessed.

There is an increasing acceptance by the general population that cannabis use is relatively benign. This understanding of cannabis risk was echoed by a U.S. survey which reported an increase over the last five years in the number of people who believe that smoking cannabis daily was safer than tobacco smoking.²⁸ The attitude change was less marked among older adults, although an increase in hospitalisation of elders with cannabis-related problems has been reported.²⁹ There are also reports of increasing use amongst perimenopausal and postmenopausal women.³⁰ This suggests that cannabis is being used by a wider range of people and increasingly seen as a safe form of self‑medication.

Currently there is a dearth of good evidence about the risks that cannabis use poses to the majority of the insured population, although such evidence is growing. Special attention needs to be given where there is current daily use, daily use as an adolescent and young adult, evidence of childhood adversities, poor education, easy access and chronic pain.

The regular use of higher potency cannabis at a young age when brain development is maximal increases the risk of developing mental health disorders, particularly schizophrenia but also cognitive problems.³¹ There are complex interactions between the drug and other mental health disorders. It remains a question whether the cannabinoids cause or relieve problems such as anxiety and depression – the answer may be both.^32,33 It also remains unclear whether heavy early use increases mortality or disability risk in later life after cessation or in the context of continued low-level recreational use.

Many people feel very strongly about these trends in cannabis use and legislation, both pro and con. This can lead to a lack of balance when commenting on the situation.³⁴ The number of studies that suggest potential cardiovascular problems associated with cannabis use is growing, but it is important to understand their context. Rarely are they controlled for route of administration and they are usually limited to those with cannabis use disorder or daily use.

Controlling for other cardiovascular risk factors significantly reduces the associations. A telephone-based study from the US reported in February 2024 assessed cannabis use in the last 30 days against self-reported cardiovascular outcomes (stroke, myocardial infarct and coronary heart disease.³⁵ Independent of age, smoking status (in this study 60% were “never smokers”), the presence of diabetes, and other risk factors, there was an independently increased risk for cardiovascular outcomes in people who used cannabis.

The authors describe the association as “strong” and “statistically significant” and probably dose-dependent. But the study was cross-sectional, not longitudinal, and relied on both honesty and accurate recall. It also did not have access to full cardiovascular risk assessments. Despite being highly suggestive, some commentators argue that the jury is still out on the cardiovascular risks posed by cannabis use.³⁶ As with all medical studies, ensuring that the studies can be generalised to the insured population is vital.^37,38

As cannabis use changes and greater understanding of the implications of regular medical and recreational use emerge, underwriting needs to be put on a clearer footing. As our understanding evolves, so must our attitude around cannabis use in the insured population. A catch-all drug misuse/abuse category does not capture the nuances of risk posed by cannabis use in society today. Never has the challenge to underwrite the individual within a consistent, evidence-based, framework of risk been more pertinent.