Show Summary Details
Page of


Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2016. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 23 January 2019

Introduction and epidemiology

Of all the psychoactive substances generally classified as illicit, cannabis (marijuana) is the most commonly used. Naturally produced cannabis comes in the form of the dried leaves and the flowering heads of the marijuana plant, Cannabis sativa. Typically, cannabis is smoked in a water-cooled pipe (a ‘bong’) or as a small hand-rolled cigarette (‘joint’). It is often combined with tobacco which increases its flammability. It can be eaten in the form of cake or cookies.

  • An estimated 160 million adults worldwide (4% of the adult population) smoke it in any one year.

  • In many Western countries, half the population aged 20–40 years report having smoked cannabis at least once in their lives, 25–30% in the past year, and 15–20% over the past month.

  • Most people who smoke cannabis do so experimentally, intermittently, or casually. Around 11% of young people become harmful or dependent users. In the EU, 1% of all adults are cannabis dependent.

  • Cannabis use is particularly common among young people, often being the first drug (apart from alcohol, tobacco, and caffeine) which they use. This raises concerns because of the vulnerability of the developing brain to its deleterious effects.

  • Cannabis contains around 100 chemicals unique to the plant which are called ‘cannabinoids’. There is a wide variation in the combination of cannabinoids in different types of cannabis.

In addition to C. sativa, other species of the hemp plant such as C. indica are smoked. Varietals and subspecies and hybrids of these two species have been developed. The C. indica derivatives often have small leaves and give off a foul odour (hence the colloquial name ‘skunk’). Cannabis grows especially well in warm, wet climates. The natural form is known as plantation or ‘bush’ cannabis. Increasingly, it is grown hydroponically, in greenhouse conditions, and this has the advantage (for producers) of producing more crops per year, and often cannabis of a higher potency.

Hashish is made from the plant’s resin. It is sold in solid pieces, and usually mixed with tobacco and smoked. Hashish oil is the most concentrated form and is typically spread on tobacco and smoked.

In recent years, cannabinoid analogues have been synthesized. More than 400 of these compounds are now recognized, and they have distinct pharmacological effects, with an especially high prevalence of psychotic phenomena described. These synthetic cannabinoids tend to be smoked by existing cannabis users. In many countries they are not classified as illicit drugs and are sold as ‘spice’. Individual compounds have particular brand names; examples are Supernova, Northern Lights, Kilimanjaro Sky, K2, Kronik, Ash, and Black Label.

Chemistry, pharmacology, and pathophysiology

  • There are more than 100 naturally occurring cannabinoids.

  • They interact with cannabis receptors, which are distributed throughout the CNS (principally CB1 receptors) and the gastrointestinal tract (mostly CB2 receptors).

  • Principal biologically active ingredient of cannabis is delta 9-tetrahydrocannabinol (∆9-THC).

  • Next commonest ingredient is typically cannabidiol, which has several effects opposite to those of ∆9-THC.

  • These two (and some other active cannabinoids) produce the desired psychological effects of cannabis smoking.

  • Δ‎9-THC produces the more vivid and hallucinatory experiences; cannabidiol causes a relaxed sensation and may have antipsychotic effects counterbalancing the effects of ∆‎9-THC.

  • An explanation for the more psychotomimetic and deleterious mental effects of new forms of cannabis (e.g. ‘skunk’) is that they contain little, if any, cannabidiol so the actions of ∆‎9-THC are not attenuated.

  • ‘Skunk’ and sinsemilla cannabis (the latter derived from the unfertilized and seedless heads of the female plant) may have a THC concentration several times that of naturally grown cannabis (up to 18%).

  • The pharmacological effects of synthetic cannabinoids are more closely related to those of ∆‎9-THC than to cannabidiol.

Many claims are made about the potency of various forms of cannabis (Table 10.1). There appears to have been a gradual increase in potency of forms of cannabis that are most commonly smoked. This may be related to varietals and hybrid forms, promotion of growth by hydroponic, greenhouse technology, and early cropping, which prevents loss of potency due to oxidation during storage. In addition, there is evidence that the typical dose per occasion has increased in recent years.

Table 10.1 Δ‎9-THC content of different preparations of cannabis

Approximate THC content (depends on the plant, soil, sunlight, humidity)

Marijuana plant:

  • leaves and stem


  • buds and heads


  • sinsemilla (unfertilized female flowering heads)


Hashish (resin from top of plant)


Hash oil (concentrated hashish extract)

15–50 %

Cannabis receptors have naturally occurring (endogenous) ligands, such as anandamide. These substances are synthesized in the CNS from membrane phospholipids and neuronal depolarization appears to influence such synthesis. The endogenous cannabinoids act on cannabis receptors (both in presynaptic and postsynaptic locations), and in doing so influence second messenger processes involved in learning and memory. The CB1 and CB2 receptors have quite different functions and locations. The main cannabis receptor is the CB1 type, which is located in areas in the brain involved with mood and memory. The psychological effects of ∆‎9-THC are mediated through CB1 receptors (the effects are lost in mice where these receptors have been deleted (knock-outs) and the effects of THC are blocked by selective CB1 receptor antagonists such as rimonabant. The gastrointestinal effects of cannabis appear to be mediated by CB2 receptors.

Absorption, distribution, and breakdown of cannabis

  • Cannabinoids are rapidly absorbed from the alveolar membrane in the lungs.

  • After smoking, plasma levels of cannabinoids are detectable in 2 minutes, with peak levels typically reached in 10–20 minutes.

  • Highly fat-soluble compounds, which distribute into body fat stores and fatty organs such as the brain.

  • Have a biphasic elimination profile: the first phase lasts about 30 minutes, and reflects redistribution of cannabinoids into fat stores.

  • The second, elimination, phase lasts 20–80 hours (sometimes more) and reflects the metabolic clearance of cannabis in the liver and excretion of water-soluble metabolites through the kidneys.

  • THC is metabolized mainly in the liver to 11-nor-THC-9-carboxylic acid.

  • Metabolites may be detected in the urine weeks after a single joint or up to 3 months after repeated daily use because of the large amount of cannabinoids that reside in body fat stores, from which they gradually leak out.

The very enduring presence of inactive cannabis metabolites is detectable in urine screens, so individuals may be found ‘drug positive’ weeks after the effects of cannabis have worn off (these generally disappear within a day). Cannabis use is thus an easy target for drug testing which in some situations (e.g. prisons) has the perverse effect of encouraging use of shorter-acting, but more dangerous drugs, such as synthetic cannabinoids, heroin, ketamine, or GHB.

Pharmacological effects

After cannabis is smoked the effects last approximately 2–4 hours, sometimes more, depending on the dose. When taken orally, the effects are delayed but may last longer, for up to 12–24 hours. The effects vary with the setting, but typically include:

  • euphoria, relaxation, sleep (in some cases anxiety or restlessness)

  • floating sensations, lightness of limbs, depersonalization

  • altered perception of time, temporal disintegration

  • rapid flow of ideas, talkativeness

  • loosening of associations, fragmented thinking

  • disturbed memory

  • conjunctival injection (red eyes)

  • tachycardia

  • elevations in blood pressure

  • increased appetite (the ‘munchies’)

  • dry mouth

  • fall in intraocular pressure

  • antiemetic effect.


PET studies show that cannabis increases blood flow to parts of the brain that mediate mood and decreases blood flow to areas associated with attention and cognition. Some of the effects of cannabis may be due to the release of dopamine (DA) in striatal and prefrontal cortical regions.

Cannabis dependence likely reflects adaptive changes in receptor function as withdrawal can be precipitated by antagonists such as rimonabant. Conversely, drugs combining THC and cannabidiol in equal proportions such as nabiximols (Sativex®) may reduce the severity of withdrawal symptoms in the short term.

Therapeutic uses of cannabis

There are several therapeutic uses of cannabinoids, for which there is evidence of efficacy. These include:

  • as an antiemetic (during chemotherapy or radiotherapy)

  • as an appetite stimulant (in HIV/AIDS patients)

  • in the relief of glaucoma

  • for spasticity and pain in certain neurological disorders, such as multiple sclerosis, spinal cord injuries, and movement disorders.

Nabilone is a synthetic cannabinoid that is licensed for the first two indications in the US and some other countries.

Currently there is a wide variety of disorders being researched (including cancer and obesity) for their responsiveness to various cannabinoids.

Medical marijuana

The term ‘medical marijuana’ is generally used to refer to the flowering heads of the marijuana plant or its crude extracts, which have not been processed pharmaceutically. These products are generally not recognized or approved as medications by regulatory authorities (such as the US Food and Drug Administration), but legislation has been passed in several countries and US states allowing individuals to smoke cannabis under medical prescription. Prescriptions are issued predominantly for pain disorders and for psychological difficulties such as anxiety and insomnia. The rationale for prescribing a smoked product is dubious.

Core clinical syndromes

Acute cannabis intoxication

This is seen after smoking a high dose of cannabis, and occasionally after seemingly small doses in susceptible people. Common features are as follows:

  • Anxiety and panic attacks: the most common adverse emotional reactions with acute intoxication. Most often reported in naïve users and more common in those with a pre-existing anxiety disorder. Such cannabis-induced panic attacks last no more than 5–8 hours, but the user may feel as if they are losing control or going mad during that time.

  • Dysphoria: there may be a period of lowered mood, which is usually mild, brief, and self-limiting.

  • Suspiciousness and paranoia: unusual when naturally grown cannabis is smoked but more common with subspecies or synthetic compounds. Can heighten the feeling of fear and loss of control. May occur in healthy individuals with no past or family history of psychotic disorders.

  • Perceptual distortions: unusual somatic or visual sensations may be experienced. These may be reported up to a week after an episode of cannabis use.

  • Visual and auditory illusions or hallucinations (may contribute to paranoid experiences).

  • Loss of insight: there may be associated paranoia; also occurs with persistent cognitive impairment.

  • Cognitive impairment: cannabis intoxication commonly results in impaired or at least altered attention, concentration, learning, and memory.

  • Psychomotor impairment: increased risk of accidents because of impaired psychomotor performance, as well as altered concentration and attention.

  • Confusion and delirium: confusion and delirium are more common with large doses of high-potency cannabis. Clinical features include confusion, persecutory delusions, hallucinations (auditory and visual), emotional lability, panic, and there may be depersonalization and derealization.

The intoxicating effects of cannabis are typically short-lived, relatively benign, and recovery is usually complete within a week of ceasing cannabis. The psychotic features such as paranoia and hallucinations may resemble some aspects of a psychotic illness, but they are transient in nature and fully resolve as intoxication clears. Perceptual distortions may occur for several weeks; they are sometimes described as flashbacks. However, because cannabis has a long half-life, the experience is not strictly speaking occurring in abstinence, so cannot be called a flashback.

Death from overdose of smoked cannabis has not been described but deaths have occurred due to violence or accidents. There is an increased risk of motor vehicle accidents, though cannabis (unlike alcohol) tends to produce inhibited, rather than disinhibited driving behaviour.

Non-dependent (hazardous/harmful) use of cannabis

Cannabis may be used on a single occasion experimentally in teenage and young adult years, or periodically at parties or other social occasions. No adverse effects are apparent with such infrequent use. Use of cannabis in the 16–30-year-old age group approaches normative in some countries where more than 50% of young people have tried it at least once. In the absence of a known threshold for harm, hazardous or ‘risky’ cannabis use is difficult to define scientifically.

Harmful cannabis use refers to repeated use of cannabis causing physical or mental harm. This can include repeated episodes of clinically significant intoxication. A common example of physical harm is recurrent bronchitis and other chest infections.

Cannabis dependence

Chronic regular daily use may extend up to 14–16 hours of continual smoking per day. The patient may be intoxicated for much of this period. A proportion of regular chronic users will exhibit features of dependence, including:

  • tolerance

  • poor control over use

  • unsuccessful attempts to stop or cut down

  • cannabis taking a higher priority over other aspects of life

  • continued despite clear evidence of harm (e.g. chest infection)

  • withdrawal symptoms on cessation of use (in some cases).

Cannabis withdrawal syndrome

The withdrawal syndrome occurs in some cannabis-dependent individuals when they cease cannabis use. The exact prevalence is not clear, but may be as high as 20% of regular heavy users. Symptoms of acute withdrawal start approximately 4 hours after cessation of cannabis, peak at 4–7 days, and last 2–3 weeks (see Box 10.1). Protracted milder withdrawals symptoms may last several weeks. Cannabis withdrawal can be precipitated by the CB1 antagonist rimonabant.

Physical and neuropsychiatric sequelae

Medical complications

These include:

  • increased risk of accidents, including motor vehicle accidents, especially if alcohol is also consumed

  • impaired pulmonary function, recurrent bronchitis, worsening of asthma, cancer of the lungs (from carcinogens in cannabis and tobacco smoke).

Neuropsychiatric complications

In addition to the acute intoxicating effects of cannabis, long-term use can lead to persistent psychiatric, neuropsychiatric, and cognitive impairment.

  • Anxiety and panic attacks: these are the most common adverse effects and occur two to four times more commonly in long-term cannabis users than in the general population.

  • Depression: following the euphoria usually experienced with cannabis intoxication, there may be a period of lowered mood, which is usually brief and self-limiting. Longer-lasting depressive symptoms are now recognized, and recent studies have established that the prevalence of depression is raised in long-term cannabis users, many of whom are cannabis dependent.

  • Paranoia: cannabis intoxication may be associated with mild levels of suspiciousness, paranoia, and loss of insight, and these can occur in healthy individuals with no past or family history of psychotic disorders. Persistent paranoia occurs in cannabis-related psychosis (see p. [link]). Paranoia is a more common acute effect of cannabis which is high in THC and low in CBD (often varieties called skunk or sinsemilla).

  • Delirium: this may occur in acute cannabis intoxication, especially when high-potency cannabis has been smoked. Clinical features include confusion, paranoid (and often persecutory) delusions, auditory and/or visual hallucinations, emotional lability, panic, and sometimes depersonalization and derealization. Recovery is usually complete within a week of stopping cannabis. Persistent delirium is described but is rare.

  • Amotivational syndrome: an ‘amotivational syndrome’ has been long described in long-term cannabis users. It comprises lack of motivation, apathy, social withdrawal, and lethargy, and there is often impaired memory and concentration. It may be due to the effects of chronic intoxication, and some dispute its existence as a discrete entity. It may improve or resolve with abstinence, unless it reflects other factors, such as personality dysfunction and other substance use.

  • Cognitive impairment: cannabis intoxication commonly results in impaired attention, concentration, learning, and memory. Regular cannabis use can also lead to often subtle cognitive impairments—in memory, attention, organization, and integration of complex information. There is usually improvement in cognitive function with cessation of cannabis use, but persistent impairment is described even with abstinence of a year or more.

Acute psychosis

There is ongoing controversy about the existence of this syndrome. Several authors have described acute psychotic episodes occurring in clear consciousness following cannabis use, and more commonly after heavy use or those with long-term use and/or cannabis dependence. The psychotic episodes are characterized by rapid onset, with a relatively benign course, and usually recovering completely within a week of abstinence even without antipsychotic agents. There is no evidence of confusion or delirium.

Psychotic features may comprise predominantly affective-like symptoms (of a manic or hypomanic type), or those resembling a schizophreniform psychosis (with auditory or visual hallucinations, delusions, and sometimes incoherent speech).

Chronic psychosis

There is also considerable debate about whether cannabis use can induce a chronic psychosis, including schizophrenia. The following might represent a consensus:

  • In patients with schizophrenia, cannabis use may trigger a relapse of the disorder and may exacerbate existing symptoms even when the patient is otherwise stable on medication.

  • Regular heavy users of cannabis may suffer repeated, short episodes of psychosis and be in a lingering, subpsychotic state; however, the psychotic symptoms will abate once cannabis use cease.

  • Cannabis use may precipitate psychotic symptoms in an individual who is in the prodrome of schizophrenia or has a high genetic or other susceptibly to it; COMT and AKT1 polymorphisms have been implicated.

  • Cohort studies indicate that cannabis users have a 2- to 4-fold increase in the relative risk for developing schizophrenia. Eliminating cannabis use in those at risk would reduce the incidence of schizophrenia by 8–15%.

  • Cannabis alone probably does not cause schizophrenia but it is an important component of a constellation of risk factors for it.

Social complications

  • Among adolescents, impaired performance at school and earlier school leaving

  • Reduced academic achievement in young people generally

  • Delinquency (in adolescence it may be difficult to distinguish the behavioural changes and moodiness of adolescence from the effects of cannabis use)

  • Impaired job performance, unemployment

  • Financial problems

  • Relationship problems, family problems

  • Criminal activity and legal problems.

Natural history

Information is scant on the natural history of various levels of cannabis use.

  • Cannabis use typically begins in the mid to late teens, and is most prevalent in early adulthood.

  • Most cannabis use is irregular, with very few users engaging in long-term daily use.

  • About 10% of those who ever use cannabis become daily users, and another 20–30% use weekly.

  • Of those who have cannabis dependence, about 50% are still smoking regularly at 5 years.

  • Smoking cannabis for up to 30–40 years is well recognized.

  • Transitions in life roles, such as entry into full-time employment, getting married, or having children, are associated with reductions in or cessation of use for many people. The largest decreases are seen in cannabis use among males and females after marriage, and especially during pregnancy and after childbirth in women.

Policy and prevention

Although cannabis is considered a less harmful drug than many, people (particularly young people) with a family history of substance dependence or mental illness such as schizophrenia or psychosis are best advised to avoid it.

Everyone should be informed about the potential neuropsychiatric complications of cannabis use.

Those using cannabis should be advised about the risks of driving while under the influence of cannabis, particularly if it is used in combination with alcohol.

There have been major global changes in regulation of cannabis in recent years. These include medical marijuana being available in 20 US states, some parts of Europe, and being fully legalized in 2014 in Uruguay and two US states.

Clinical assessment and diagnosis


As part of the history of substance use (see pp. [link][link]), enquire about the three dimensions of cannabis involvement, as follows:


  • What is smoked? Cannabis leaf (low potency), the more potent flowering heads (‘bud’)?

  • How is it produced? Naturally grown, hydroponically grown?

  • Or are particular forms smoked (e.g. skunk, sinsemilla) or synthetics?

  • How is it smoked? As a joint, or using a bong or pipe?

  • How much is smoked per occasion/per day? Quantify in ‘cones’, grams, or ounces: 1 ounce = 28 grams; 1 gram = 15 cones.

  • How many hours in a day do you feel the effects (are ‘stoned’)?

  • How frequently is it smoked? Daily, or episodically?

  • Is there a pattern to its use, e.g. at weekends only?

  • What is the duration of smoking? Months, years, decades?

  • When was the last smoke?


  • Do you experience withdrawal symptoms (indicates dependence)?

  • Have you continued to smoke cannabis despite problems?

  • Is cannabis use taking over your life?

  • Are you affected (‘stoned’) for more than 4–6 hours per day regularly? (Suggests dependence).


  • History of respiratory illnesses.

  • History of psychiatric illness and possible relationship to cannabis use.

  • Social, occupational, and legal problems.

Taking a family history of cannabis (and other substance) use and a family history of psychiatric illness are also important.


Clinical examination may be normal, but the following may give clues to cannabis involvement:

  • Conjunctival injection (red eyes).

  • Tachycardia, raised blood pressure.

  • Signs of chronic obstructive airways disease.

Mental state examination may reveal anxiety, panic, depression, confusion, paranoid ideation, and (in heavy users/ those with cannabis dependence), features of psychosis (see pp. [link][link]).

Laboratory investigations

Cannabis is lipophilic and can be detected in plasma for up to a month. In some countries, blood and/or saliva are tested to determine competence to drive.

Brief interventions

Cannabis users who are non-dependent, but smoking cannabis in a hazardous or harmful way can be offered a brief intervention based on the FLAGS acronym:

  • Feedback: on any medical, neuropsychiatric, or social harms experienced as the result of cannabis use.

  • Listen: to the patient’s response—does the patient want to quit using?

  • Advice: convey clear medical advice, e.g. the potential medical, neuropsychiatric, and social complications of cannabis use and the benefits of not smoking cannabis (important particularly for young people).

  • Goals: set goals tailored according to the individual patient’s response.

  • Strategies: set out strategies to achieve the goals. Offer follow-up to determine progress and offer support.

Management of intoxication and withdrawal

Cannabis intoxication

Intoxication should be managed conservatively in a quiet environment with reassurance and food intake (e.g. chocolate, coffee).

Withdrawal management (detoxification)

Most people who smoke cannabis can stop without the need for medical involvement. However, in those who have developed cannabis dependence, the occurrence of a withdrawal syndrome often stops attempts at ceasing within a few days. The cannabis withdrawal syndrome is an unpleasant experience, particularly as it tends to increase in severity over the first few days (peaking at 6–8 days) and it may be more than a month by the time it resolves. Providing information about the nature of withdrawal symptoms and their time course is a key component of intervention. Standard psychological interventions such as motivational interviewing and behavioural rescheduling are helpful.

Medication is often necessary to manage the withdrawal syndrome. The most effective agents are cannabinoid agonists e.g. nabiximols (Sativex®). Otherwise, the evidence base for withdrawal management is small. Various treatments are used in clinical practice:

  • Benzodiazepines (e.g. diazepam 5–10 mg every 6 hours for 7–10 days and ceasing by 14 days) improve insomnia but do not consistently relieve other withdrawal symptoms.

  • A low dose of a sedating antipsychotic agent (e.g. quetiapine 50–100 mg as the extended-release form mid evening) may be prescribed to assist sleep and can improve appetite; no improvement in mood is reported. The patient should be warned that the first dose might cause significant sedation and to contact the prescriber if that occurs.

  • Mirtazapine 15–30 mg at night also helps with insomnia, but does not appreciably relieve dysphoria.

Management of cannabis use and dependence

There is no known dividing line between safety and harm for cannabis use, and so the level of use at which some intervention is appropriate is not established. For people with likely hazardous cannabis use or harmful use, a brief intervention is appropriate (see p. [link] and p. [link]). For patients with cannabis dependence and/or established cannabis-related disorders, more formal management is appropriate. Indeed, cannabis dependence is the primary reason for 28% of new referrals to drug services in the EU, second only to heroin (41%). The key components of management are listed in Box 10.2.

Management of complications

Management of acute mental disorders

Mostly this entails confirming the diagnosis of a cannabis-related disorder, and reassuring the patient (where this is possible) that this is the cause. Acute mental disorders will usually start to abate in 5–8 hours.

  • For acute anxiety, a benzodiazepine may be prescribed in the short term.

  • Depressive symptoms will need to be assessed as to the likely cause, whether an underlying mood disorder, a cannabis-induced syndrome, or symptomatic of another disorder. Detoxification and review is important prior to considering the place of antidepressants or therapy.

  • Psychotic disorders will also need to be assessed as to likely cause and relationship with cannabis use. Detoxification is necessary, combined with appropriate antipsychotic medication.

Management of persisting mental disorders

Psychosis associated with cannabis use can be managed by reducing level of use over time. Harm reduction may be achieved in those not able to reduce by switching from higher THC brands to lower, especially those higher in CBD. In some cases, antipsychotics may be indicated.

Further reading

Advisory Council on the Misuse of Drugs (2008). Cannabis: Classification and Public Health. London: UK Home Office.Find this resource:

Allsop DJ, Copeland J, Lintzeris N, et al. (2014). Nabiximols as an agonist replacement therapy during cannabis withdrawal. JAMA Psychiatry 71(3):281–91.Find this resource:

Budney A, Hughes A. (2006). The cannabis withdrawal syndrome. Curr Opin Psychiatry 19:33–238.Find this resource:

Curran HV, Morgan CJA. (2014). Desired and undesired effects of cannabis on the human mind and psychological well-being. In Pertwee R (ed) Handbook of Cannabis, pp. 647–60. Oxford: Oxford University Press.Find this resource:

Moore THM, Zammit S, Lingford-Hughes A, et al. (2007). Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. Lancet 370:319–28.Find this resource:

Morgan CJA , Duffin S, Hunt S, et al. (2012). Neurocognitive function and schizophrenia-proneness in individuals dependent on ketamine, on high potency cannabis (‘skunk’) or on cocaine. Pharmacopsychiatry 45:269–74.Find this resource:

Morgan CJA, Schafer G, Freeman TP, et al. (2010). Impact of cannabidiol on the acute memory and psychotomimetic effects of smoked cannabis: naturalistic study. Br J Psychiatry 197:285–90.Find this resource:

Nutt DJ, Nash J. (2002). Cannabis—An Update. London: Home Office Publications. Available at: this resource:

Rawlins M. (2008). Cannabis: Classification and Public Health. London: Home Office Publication. Available at: