Terminology

The literature and diagnostic classifications such as the DSM and ICD use varying terminology when describing substance-use disorders. Here we differentiate between abuse (taking a drug to achieve an appetitive effect, or ‘high') and misuse (any use that deviates from the way a medication has been prescribed).

A reminder of what the evidence tells us

The bulk of scientific literature on benzodiazepine safety, dependence and misuse tells a different story. Although demonstrating a range of potential liabilities, including cognitive and psychomotor impairment, possible risk in pregnancy and severe and/or prolonged withdrawal syndromes, it does not confirm that these medications are primary drugs of abuse or gateway drugs leading to other substance abuse. The database was scrutinised in the 1980s and 1990s in a series of extensive reviews, including a volume commissioned and published by the American Psychiatric Association. In aggregate, they comprise over 2000 literature citations, dealing with both animal and human studies bearing on abuse, misuse and dangerousness of benzodiazepines.^{Reference Woods, Katz and Winger1–Reference Woods, Katz and Winger3} Their authors conclude that benzodiazepines ‘do not strongly reinforce their own use and are not widely abused drugs. When abuse does occur, it is almost always among persons who are also abusing alcohol, opiates or other sedative hypnotics’² and that ‘epidemiological studies of various populations of drug abusers have often found rates of nonmedical use of benzodiazepines that exceed those found in the general population [but] the preponderance of the extensive use of benzodiazepines is directed by physicians for disorders in which these drugs have proven therapeutic effect’.^{Reference Woods, Katz and Winger3} Although co-abuse of benzodiazepines has risen in the context of the opioid epidemic, there has been no newer evidence suggesting that either benzodiazepine abuse or any other substance abuse has its genesis in prescribed treatment for general (i.e. non-substance-abusing) patients. In his 2005 review of benzodiazepine abuse and dependence, O'Brien states, ‘benzodiazepines are usually a secondary drug of abuse – used mainly to augment the high received from another drug or to offset the adverse effects of other drugs. Few cases of addiction arise from the legitimate use of benzodiazepines’.^{Reference O'Brien4} Although most of the literature on this topic is not recent, neither is it outdated; it is simply ignored.

Reasons for the bias against benzodiazepines

Why bias against a safe and useful class of medications has become so entrenched is not entirely clear and is itself a subject worthy of investigation. One factor may be that major pharmaceutical companies long ago abandoned benzodiazepines in favour of antidepressants, which we believe has had a substantial influence on practitioners and has left benzodiazepines with few people to speak up for them. Concern about co-abuse of benzodiazepines by opioid abusers, with potentially lethal consequences, may be another factor motivating physicians to avoid them. In this climate of opinion, discussions about benzodiazepines often blur important distinctions about their clinical pharmacology or describe them in inaccurately pejorative terms. We discuss below five instances of unfounded beliefs about benzodiazepines that we believe have been especially detrimental.

1. (a) Benzodiazepines prescribed for anxiety disorders are likely to be abused. Benzodiazepines have a short latency of onset to calming or sedating effects, which may make them attractive to people who abuse substances. However, they are not prone to being abused by those with no such history. Conflating risk in these two populations stigmatises people with anxiety disorders and deprives them of treatment that might restore them to more functional lives.

2. (b) Patients who misuse benzodiazepines are on a spectrum of drug abuse and are at risk of proceeding to frank abuse or addiction. Misuse is defined as any use of a medication that deviates from the way it has been prescribed by a clinician – including taking extra doses or taking less medication than prescribed.^{Reference Maust, Lin and Blow5} Addiction (called dependence in ICD-10) is defined by a cluster of behaviours that includes drug-taking to achieve appetitive effects (i.e. a high), preoccupation with the substance in question, temporary satiation, loss of control and persistent use despite negative consequences. Misuse of benzodiazepines is common, estimated at 17% of overall use.^{Reference Maust, Lin and Blow5} However, the great majority of people who deviate from doctors’ prescriptions of them are trying to control symptomatic distress, not to get high,^{Reference Maust, Lin and Blow5} and there is no evidence that misuse is likely to lead to abuse. Abuse and addiction should be addressed by substance abuse treatment; misuse is a more heterogeneous phenomenon that may involve suboptimal prescribing, poor doctor–patient communication, and patients inappropriately attempting to eradicate all negative affect with medication. However, alarmed clinicians who automatically view patients who deviate from their instructions as medication abusers may demand that they taper off benzodiazepines, with detrimental consequences.

3. (c) Patients prescribed benzodiazepines tend to escalate their doses, which should preclude long-term use. There is a common belief that long-term benzodiazepine treatment is associated with tolerance to their anxiolytic effects and consequent dose escalation. The accumulated evidence is to the contrary: long-term treatment is associated with maintenance of therapeutic benefits, and no dose escalation. Tolerance does develop to the sedating and psychomotor effects of benzodiazepines, however.² Failure to make this distinction may be the basis for withholding benzodiazepines, or for withdrawing them from patients who have been doing well on them during acute treatment.

4. (d) Benzodiazepines are dangerous in overdose. Benzodiazepines alone are among the safest of psychotropic medications, with lethal dose LD₅₀ estimates for most in the range of thousands of mg/kg. Even alprazolam, which may be more toxic, has an estimated LD₅₀ range of 300–2000 mg/kg. Taken in conjunction with alcohol or opioids, they markedly raise the lethality of these already dangerous substances. That benzodiazepines are safe for the vast majority of people with anxiety disorders for whom they are prescribed is obscured by commonly used phrases such as ‘benzodiazepine-related death’ to describe a lethal combination of opioids and benzodiazepines ingested by a polysubstance-abusing person.

5. (e) Taking benzodiazepines long-term leads to dependence. The word ‘dependence’ almost invariably has pejorative connotation and may unfairly characterise patients when applied vaguely or inconsistently. As used in ICD-10, ‘drug dependence’ is essentially a syndrome of addiction. ‘Dependence’ may also be used to describe a physiological withdrawal syndrome, an entirely different phenomenon that occurs with many medications, and is not in itself a sign of addiction.^{Reference O'Brien4} People with anxiety disorders discontinuing chronic benzodiazepines may experience a syndrome that includes rebound anxiety, which clinicians may take as a reason to withhold long-term treatment. Dependence is commonly applied, without clear definition, to patients in long-term treatment with benzodiazepines who lack any of the behavioural characteristics of substance abuse. They may be labelled as dependent (or addicted or hooked) because of the potential for a withdrawal syndrome and told that they must deal with it by getting off their medication. It is not surprising that anxious patients summarily told that they are drug dependent and deprived of an effective medication have difficulty tapering off it. But their doctors may interpret their struggles as evidence that a benzodiazepine prescription was problematic to begin with.

A call to clinicians

It's time to return to the evidence about benzodiazepines and to conceptual rigour in interpreting it. Benzodiazepines are highly effective for treatment of anxiety disorders, but are not for everyone, have potential liabilities and are best used in conjunction with targeted psychotherapies. That polysubstance abuse often includes benzodiazepines, however, should not blind us to their appropriate use. Distinctions between abuse, addiction/dependence, misuse and physiological dependence may be challenging, but they are supported by the evidence and are clinically important. Conflating these phenomena will perpetuate stigma against benzodiazepines, the clinicians who prescribe them and the patients who take them. We invite colleagues to engage in evidence-based reappraisal of the benefits and risks of these medications and to abandon aspects of conventional wisdom that do not stand up to such scrutiny.

Acknowledgement

This manuscript arose out of discussions among the members of the International Taskforce on Benzodiazepines, an informal collaborative group of academic clinicians (including the authors) who are interested in disseminating accurate information about and fostering appropriate use of these medications.