Evidence-based guidance for benzodiazepines

Sleep disturbance is the third most common psychological reason for patient encounters in general practice. Population surveys found 13–33% of adult Australians have regular difficulties getting to sleep or staying asleep.¹⁰⁴

Insomnia is defined in DSM-5 as a difficulty in getting to sleep, staying asleep or having non-restorative sleep despite having adequate opportunity for sleep, together with associated impairment of daytime functioning, with symptoms being present for at least 4 weeks.⁴⁰

Acute insomnia meets the DSM-5 definition of insomnia, but with symptoms occurring for less than 4 weeks.⁴⁰ It is experienced by up to 80% of the population at some stage, generally due to one or more precipitating factors.¹⁰⁵

Factors that can precipitate acute insomnia include:¹⁰⁴

• physiological – hyperarousal due to stress, being ‘on-call’, caring for a sick child/relative, being in a strange situation (eg in hospital)
• pharmacological – prescribed drugs (eg newly prescribed diuretic causing nocturia) and non-prescribed drugs
• physical – coughing, environment (eg noise, temperature)
• disruption of circadian rhythm (eg jet lag).

Provided patients adhere to good sleep habits, most will return to normal sleep once the precipitating factor has resolved or diminished.¹⁰⁶

Chronic insomnia was previously viewed as a sleep disturbance that was secondary to a medical condition, psychiatric illness, sleep disorder or medication, and would improve with treatment of the underlying disorder.^107,108 However, evidence over the last 20 years indicates this view is incorrect. It is now recognised that insomnia may be an independent disorder.^108–110 The understanding of chronic insomnia continues to evolve. Insomnia may have some similarities with depression in that they both represent long-term disorders for which many patients require maintenance treatment.

Chronic insomnia may occur in the absence of coexisting conditions. When coexisting conditions exist, insomnia may persist despite successful treatment of the coexisting condition. Treatment directed at the insomnia, rather than the comorbidity, may be necessary. Since insomnia can precipitate, exacerbate or prolong comorbid conditions, treatment of insomnia may in turn improve comorbidities.

Chronic insomnia is unlikely to resolve spontaneously.¹¹¹

Management of insomnia in general practice

When patients present with insomnia to GPs in Australia, few (0.8%) receive a referral for specific non-drug therapy.¹¹² All guidelines strongly recommend psychological and behavioural management.^{68,102,108,113}

Assessment and diagnosis

Assessment and diagnosis of insomnia requires:¹¹¹

• understanding the patient’s typical sleep pattern over an extended time frame (weeks to months) – a sleep diary can help assessment
• identifying contributing lifestyle factors (eg caffeine, nicotine, pets in the bedroom)
• understanding the patient’s beliefs and concerns about sleep
• determining the effects of poor sleep on the patient (eg poor memory, fatigue, work absence, accidents)
• identifying comorbid conditions – this may be aided by the Auckland Sleep Questionnaire, which is a validated screening tool.¹¹⁴

Acute insomnia

Treatment is focused on avoiding or withdrawing the precipitant, if possible. All patients should receive basic behavioural counselling on sleep hygiene and stimulus control.

Drug treatment may be indicated as an adjuvant to non-drug therapies for acute insomnia that is severe, disabling and causing distress. Benzodiazepines and Z drugs are the most effective drugs.²⁵ The short-term use of benzodiazepines as hypnotic agents should only be one aspect of general management,²⁵ with a clear endpoint of drug cessation once sleep patterns return to normal.

Chronic insomnia

Management of chronic insomnia starts with addressing any relevant, underlying problems that are present, such as:

• pharmacological – prescribed drugs (eg some antidepressants, withdrawal of sedatives) and non-prescribed drugs (eg caffeine, alcohol)
• physical (eg pain, respiratory and cardiovascular disorders, neurological disorders, movement disorders, restless leg syndrome and other sleep disorders)
• psychiatric disorders (eg depression, anxiety, dementia and substance misuse)
• disruption of circadian rhythm (eg shift work).

Non-drug interventions

First-line therapy for chronic insomnia should be non-drug interventions that are supported by evidence in achieving sustained improvements in sleep parameters.^115,116 Interventions include:

• CBTs^18,117 (eg stimulus control, sleep restriction therapy, relaxation techniques, cognitive therapy and sleep hygiene education)¹¹¹
• brief behavioural therapy (ie modification of waking behaviours that affect the physiological systems regulating sleep)^118–121
• exercise.¹²²

Psychological and behavioural treatments administered weekly over a 4–8– week period have shown robust and stable improvements in sleep continuity for up to 2 years.¹²³ These therapies are now available online, which has vastly improved access.

Refer to Resource G for a GP guide to behavioural therapies for insomnia.

Drug therapy

For patients who continue to have insomnia that is sufficiently burdensome to warrant other interventions, reasonable approaches include continued behavioural therapies, medication or both.¹⁰⁸

Recommendations in guideline on the use of medication for chronic insomnia vary. This is primarily due to the lack of evidence from extended trials that adequately compare long-term risks and harms of these medications. Some guidelines suggest avoidance of hypnotic drugs as far as possible,¹⁰² while others make a considered judgement on the risks and benefits.^68,108,113

The decision to treat chronic insomnia must weigh any potential serious side effects associated with pharmacologic therapy against potential health risks of not providing treatment. This includes decreased quality of life, increased risk for psychiatric comorbidities and SUD, and decreased performance. The approach should be individualised according to the patient’s values and preferences, the availability of advanced behavioural therapies, the severity and impact of the insomnia, and the potential benefits versus the risks, costs and inconveniences.¹⁰⁸

Pharmacological treatment should be accompanied by specific patient education, regular review and continued effort to employ the lowest effective maintenance dosage of medication, and to taper medication when conditions allow.¹¹³

• Benzodiazepines – May have a place in the treatment of severe acute insomnia or treatment-resistant chronic insomnia. For patients with sleep onset insomnia, a short-acting medication is a reasonable choice for an initial trial of pharmacologic therapy. This may improve the insomnia with less residual somnolence the following morning. For patients with sleep maintenance insomnia, a longer-acting medication is preferable for an initial trial of pharmacological therapy.¹⁰⁸
• Z drugs – Have been shown to be effective in the treatment of insomnia in the short term. However, there is very limited evidence that they retain their efficacy during long-term treatment.¹²³ Z drugs are associated with issues relating to adverse events, rebound insomnia, development of tolerance and SUD, and need to be carefully monitored in individual patients.¹²³ Zolpidem’s black box warning states, ‘Zolpidem may be associated with potentially dangerous complex sleep-related behaviours which may include sleep walking, sleep driving and other bizarre behaviours. Zolpidem is not to be taken with alcohol. Caution is needed with other CNS depressant drugs. Limit use to four weeks maximum under close medical supervision’.¹²⁴
  Psychological and behavioural treatments produce comparable outcomes to Z drugs during active treatment and have better durability beyond the active administration of treatment.¹²³
  Prescription of zolpidem and zopiclone should be treated with the same precautions and patient management as benzodiazepines.¹⁰²
• Melatonin – An endogenous hormone associated with the control of circadian rhythms and sleep regulation. Melatonin levels may be reduced in middle-aged and elderly patients with insomnia. Supplementation with melatonin has been shown to improve limited aspects of sleep in 30–50% of patients over 55 years of age.^106,125,126 Data from clinical trials are variable. Some showed that some patients gain clinically significant improvements in quality of sleep and morning alertness with prolonged-release melatonin, but many patients in the clinical trials did not respond to treatment.³⁵ Melatonin does not appear to be addictive or cause withdrawal effects when stopped.¹²⁵ There are a lack of long-term study data.

   

  At the time of writing, melatonin is available in Australia as a prolonged-release formulation for short-term treatment of primary insomnia, characterised by poor sleep quality in patients aged 55 years or older.³⁵ The recommended dose is 2 mg orally at bedtime, and at present, there is insufficient evidence to support treatment beyond 3 weeks.³⁵
  In a European campaign to reduce benzodiazepine and Z drug usage, the availability of prolonged-release melatonin was shown to positively contribute to success.¹²⁷

• Other prescription medications – Includes antipsychotics (eg quetiapine, olanzepine). Studies demonstrating the usefulness of these medications for either short- or long-term management of insomnia are lacking. Furthermore, these agents have significant risks and therefore, their use in the treatment of chronic insomnia (without a relevant comorbid condition) cannot be recommended.¹⁰⁷ Numerous other medications have a sedating effect, but are not recommended for routine use in patients with insomnia. These include antidepressants, diphenhydramine and antipsychotics.¹⁰⁸
• Over-the-counter (OTC) medications – Such as antihistamines or antihistamine/analgesic type drugs, and herbal and nutritional substances (eg valerian) are not recommended in the treatment of chronic insomnia due to the relative lack of efficacy and safety data.

Resources

• Medical Journal of Australia supplement, Sleep Disorders: A practical guide for Australian health care practitioners
• Australasian Sleep Association, Insomnia
• Australian Prescriber, Cognitive behaviour therapy in medical practice, available at www.australianprescriber.com/magazine/24/2/33/7
• Australian Psychological Society, Managing insomnia: What we’ve learnt in the last 10 years, available at www.psychology.org.au/inpsych/2014/april/bartlett
• RACGP Handbook of non-drug interventions (HANDI), Brief behavioural therapy for insomnia in adults
• South Australian kit – Insomnia management kit

Background

Anxiety disorders include GAD, panic disorder, OCD, PTSD, phobias and SAD.

In 2007, anxiety disorders were the most common self-reported mental disorder in Australia, affecting 14% of people aged 16–85 years.¹²⁹ Anxiety disorders may be prominent in depressive conditions and other chronic health diseases. Anxiety and related disorders often become chronic.

Not all patients with anxiety symptoms require treatment. Anxiety symptoms exist on a continuum and many people with milder degrees of anxiety, particularly recent onset and association with stressful situations will recover without intervention.⁸⁹ The need for treatment is determined by the severity and persistence of symptoms, the presence of comorbid mental or physical illness, the level of disability and the impact on social functioning. Treatment should aim to achieve full remission of symptoms and return of function, rather than just symptom improvement and distress reduction.¹³⁰

Randomised controlled trials across a range of anxiety disorders often demonstrate a high placebo response, which indicates that non-specific effects can play a large part in improvement.¹²⁸

Approaching anxiety disorders systematically involves identifying and treating any comorbidities, providing patient education and appropriate psychological and pharmacological interventions. These should be evidence-based and patients should receive ongoing monitoring to determine whether treatment aims are being achieved.¹³⁰

Management of anxiety in general practice

For a comprehensive review of management of anxiety disorders, GPs are advised to review individual clinical guidelines.

Benzodiazepine use in anxiety disorders is mostly limited to severe, or treatment-resistant, cases. Patients with a history of significant mental illness who use two or more psychoactive drugs (polydrug use) may be more vulnerable to major harms. Significant caution should be taken if prescribing benzodiazepines to patients with comorbid alcohol or SUDs, or polydrug use. GPs should consider seeking specialist opinion in the management of these patients.

Assessment and diagnosis

Comprehensive clinical assessment is the first step to developing a diagnosis and determining the patient’s level of disability. In milder, recent onset anxiety disorders, consider ‘watchful waiting’ (support, addressing social factors and monitoring).⁸⁹

It is important to detect comorbid depression. Depression should be treated if depressive symptoms are moderate or severe.

Most guidelines recommend CBT as first-line non-drug therapy, while SSRIs and SNRIs are the drugs of first choice. Benzodiazepine recommendations are generally limited to severe or treatment-resistant cases. However, the efficacy of psychological and pharmacological approaches is similar in the acute treatment of mild to moderate anxiety disorders.^89,94,131

The selection of an initial treatment modality should be guided by considerations including the patient’s needs and preferences, the risks and benefits for the patient, the patient’s past treatment history, the presence of comorbid general medical and other psychiatric conditions, cost and the local availability of evidence-based psychological interventions.

Where appropriate and available, patients should be offered a choice of evidence-based treatment approaches.

Cognitive behavioural therapy

All major guidelines recommend CBT as the first-line intervention for anxiety disorders.^64,89,94,130

CBT is a multimodal intervention. Specific techniques used in the therapy include education, self-monitoring, relaxation training, cognitive restructuring, exposure to imagery and anxiety-producing situations, and relapse prevention. CBT has been shown to be an effective stand-alone treatment for GAD.¹³² Comorbidity does not decrease the treatment effects of CBT.

CBT for most anxiety, and related disorders, can be delivered effectively in individual or group therapy formats.¹³⁰ There are also an increasing number of self-directed formats that require minimal or no therapist contact, which have been shown to be effective.¹³⁰ These include bibliotherapy (self-help books) and internet or computer-based programs.^133,134

A combination of medication and cognitive behaviour or exposure therapy has been shown to be a clinically desired treatment strategy.⁹⁴ However, combination therapy results have been conflicting,^135,136 and results vary for different anxiety disorders. While current evidence does not support the routine combination of CBT and pharmacotherapy as initial treatment for all anxiety disorders, there is support for combined use in panic disorders, with or without agoraphobia.^137,138

Benzodiazepines are generally avoided in patients with anxiety disorders who are undergoing CBT. This is due to their potential interference with motivation and learning, which are required for CBT to be effective. Some authors are now challenging this,⁷³ however there is sparse trial evidence to support a conclusion. More research is needed to ascertain if these treatment modalities can be combined effectively.

CBT protocols for anxiety usually involve 10–14 weekly sessions, but briefer strategies of 6–7 sessions have been shown to be as effective. Unfortunately, a lack of access to trained clinicians may be an issue in some areas and therefore lead to the majority of patients with anxiety being treated with medications.¹³⁹ Online CBT programs have shown efficacy, and may be suitable for patients who cannot access face-to-face therapy, or who prefer treatment in their own homes, in their own time.

Antidepressants

Note that anxiety disorders show a strong placebo response, especially at mild to moderate levels of symptom severity.⁸⁹ If pharmacotherapy is indicated, the SSRIs and SNRIs are preferred agents.¹⁴⁰ Tricylic antidepressants (TCAs) and monoamine oxidase inhibitors (MOAI) are other alternatives.

Although preferred over benzodiazepines, there are limited studies comparing head-to-head effectiveness with antidepressants. Reviews of the studies performed suggest comparative effectiveness of benzodiazepines to older and new antidepressants.^141–142

In trials of benzodiazepines and newer antidepressants, benzodiazepines have demonstrated comparable or greater improvements with fewer adverse events in patients suffering from GAD or panic disorder.¹⁴¹ Efficacy of benzodiazepines for panic disorder is comparable to SSRIs, SNRIs and TCAs.¹⁴³ Similarly, the incidence of withdrawal symptoms from antidepressants seems to occur at similar levels to benzodiazepines.⁹⁶

Reviewers have suggested the major change in prescribing pattern from benzodiazepines to newer antidepressants in anxiety disorders has occurred in the absence of comparative data of high-level of proof.^141,144

However, SSRIs and SNRIs remain recommended first-line treatments by international guidelines for anxiety disorders.¹⁴³

Short-term benzodiazepines as occasional adjunctive therapy may be effective at reducing worsening of symptoms that can occur in the first days to weeks of initiating antidepressant medication.

Benzodiazepines

Benzodiazepines are not indicated for ‘mild’ anxiety.

Benzodiazepines may be used (as monotherapy or in combination with antidepressants) for patients with very distressing or impairing symptoms whom rapid symptom control is critical.¹⁴³

Benzodiazepines have evidence of benefit for GAD, social anxiety disorder and panic disorder, but not for OCD or PTSD.^64,94 Trials have been conducted with clonazepam, diazepam and lorazepam, which have demonstrated the efficacy of these compounds in managing panic disorder clonazepam for SAD, diazepam and bromazepam for GAD.³

The benefit of a more rapid response to benzodiazepines must be balanced against the possibilities of troublesome side effects (eg sedation) and physiological dependence that may lead to difficulty discontinuing the medication.¹⁴³ Note that:

• Due to its rapid onset and offset of action, alprazolam is the benzodiazepine most commonly prescribed for panic disorders. However, in a meta-analysis, it has not been shown to have better efficacy than other benzodiazepines for panic disorders, and it does have a greater risk of dependence, problematic use and withdrawal.¹⁴⁵
• Although tolerance is less of an issue with anxiety, patients are at risk of dependence and other harms (eg depression, increased anxiety, accidents).

When benzodiazepines are prescribed short term for severe anxiety, they are generally used in conjunction with other interventions including counselling or antidepressants (where appropriate), to reduce the risk of symptom recurrence⁸⁹ or to alleviate and prevent the worsening of anxiety¹³⁰ that may occur at the start of antidepressant therapy.¹⁴⁶

Rarely, ongoing therapy with benzodiazepines may be necessary in patients with severe, treatment-resistant anxiety. Although concerns have surrounded the risks of tolerance and SUD with long-term use of benzodiazepines, there is little evidence of tolerance to their anxiolytic effects.³⁵ Problematic use is a risk in those with a history of SUD, but is otherwise uncommon.¹⁴⁷

The decision to treat chronic anxiety with benzodiazepines must weigh the risks and benefits of benzodiazepine therapy. Concerns about potential problems in long-term use should not prevent their use in patients with persistent, severe, distressing and impairing anxiety symptoms,¹⁴⁷ or in patients who are resistant to, or cannot tolerate, multiple first-line therapies.¹³⁰ Ongoing supervision is required.

Resources

• Australian Prescriber, Cognitive behaviour therapy in medical practice
• RACGP Handbook of non-drug interventions (HANDI), Depression and anxiety: internet based or computerised CBT (iCBT or CCBT), includes relevant website links

Background

Alcohol dependence is characterised by craving, tolerance, preoccupation with alcohol and continued drinking in spite of harmful consequences (eg liver disease). From a clinical perspective, alcohol dependence can be subdivided into mild, moderate or severe categories. People with mild dependence (based on alcohol dependence assessment tools) usually do not need assisted alcohol withdrawal. People with moderate dependence usually need assisted withdrawal, which can be managed in a community setting. People who are severely alcohol dependent need assisted alcohol withdrawal, typically in an in-patient or residential setting.¹⁴⁹

The onset of alcohol withdrawal is usually 6–24 hours after the last drink. In some severely dependent drinkers, withdrawal can occur while the patient is still intoxicated due to drops in blood alcohol level.³⁰

About 95% of people with alcohol dependence can stop drinking without major withdrawal symptoms (ie delirium, seizures).³⁰ Supportive care alone is often effective in minor alcohol withdrawal. Medication may not be necessary with lower daily alcohol consumption or for periodic drinkers. Seizures (usually a single episode) occur in approximately 5% of people with alcohol dependence when withdrawing from alcohol, and delirium tremens are typically seen in the most severe forms of alcohol withdrawal.³⁰

Management of alcohol withdrawal in general practice

Alcohol dependence may require complex multidisciplinary therapy, including social support systems.

Assessment

Formal assessment tools can help determine if the patient is best suited to alcohol withdrawal in the primary care or specialist alcohol services setting.

The routine screening tool is the Alcohol Use Disorders Identification Tool (AUDIT or AUDIT-C) (refer to Resource F in the PDF version). Other tools include Severity of Alcohol Dependence Questionnaire (SADQ) to assess the severity of dependence; and the Clinical Institute Withdrawal Assessment of Alcohol State, revised (CIWA-Ar) or Alcohol Withdrawal Scale (AWS) for severity of withdrawal.

Patients may be suitable for home alcohol withdrawal if they:

• have no history of seizures or delirium tremens
• do not pose a suicide risk
• have social support
• show no significant polydrug misuse
• are not dependent on benzodiazepines.¹⁴⁹

Assisted withdrawal approaches

When conducting an assisted withdrawal from alcohol, there are three commonly used approaches:³⁰

• benzodiazepine loading – involves giving a large dose (up to 80 mg of diazepam) on day one in an in-patient setting, and then no further benzodiazepines
• tapering dose regimes – a predetermined (fixed) dose of benzodiazepine is administered in tapering doses over 2–6 days
• symptom-triggered sedation – doses of benzodiazepine are administered according to the severity of withdrawal symptoms.

There is limited evidence for selecting one approach over another, however, the first approach is more suited to an in-patient setting. In a comparison of fixed schedule and symptom triggered regimens, the latter two approaches were favoured.¹⁵⁰

Benzodiazepines

Benzodiazepines have been shown to be one of the most effective drug classes in the management of alcohol withdrawal syndrome. The rationale of the use of benzodiazepines is to modulate CNS hyperactivity due to the alcohol withdrawal, by interacting with GABA receptors.¹⁵⁰

A 2010 systematic review found benzodiazepines have a protective benefit against alcohol withdrawal symptoms, in particular seizures (when compared to placebo), and have a potentially protective benefit for many outcomes when compared with other drugs. This review did not find a statistical significance among the performance of different benzodiazepines.¹⁵⁰

Avoid prescribing benzodiazepines to patients who are dependent on alcohol before enrolling them in, or referring them to, community withdrawal. Prescribing a ‘small dose of diazepam’ to help patients who claim to have stopped or reduced their alcohol consumption should also be avoided. Benzodiazepines can increase alcohol cravings and relapse rates.¹⁵¹

Baclofen and carbamazepine

Baclofen has been used to rapidly reduce symptoms of severe alcohol withdrawal syndrome. However, the evidence for recommending baclofen is insufficient to support its use in most situations.¹⁵² Carbamazepine has been effective in trials, but its role is as yet unclear.¹⁵³ These, and several other drugs, may be useful adjuncts but cannot currently be recommended for clinical monotherapy.

Background

Bipolar disorder is a recurrent, disabling condition. Patients experience periods of mania or hypomania, depression and mixed episodes or ‘dysphoric mania’ (both manic and depressive symptoms). Bipolar disorder is commonly subdivided into bipolar disorder I (at least one manic episode) and bipolar disorder II (hypomania and depression only).¹⁵⁴

Bipolar disorder has a lifetime prevalence of up to 1.6%.¹⁵⁴

Management of manic episodes in general practice

For acutely manic patients, referral to a specialist psychiatric service for in- or out-patient care is recommended.¹⁵⁴

There are two components to managing acute mania with medication:¹⁵⁴

• mood stabilisation with drugs (eg lithium, sodium valproate, carbamazepine) or one of the new second-generation antipsychotics
• adjunct therapy to control acute agitation while mood stabilisers start to take effect (about 1 week for most patients).

Adjunct therapy usually consists of an antipsychotic or benzodiazepine (or a combination). This calms or sedates the person with mania as a temporary measure, until the mood stabiliser starts to be effective.¹⁵⁴

In most cases, the benzodiazepine or antipsychotic would be withdrawn once the acute episode has resolved, and then only the mood stabiliser continued. It is common practice to taper down and discontinue benzodiazepines within 2–3 weeks of achieving adequate symptom control in mania.¹⁵⁴

Background

Epilepsy affects approximately 224,000 Australians (1 in 100) and is the most common chronic brain disorder. Nearly 10% of Australians will have a seizure during their lifetime, and one-third of those will be diagnosed with epilepsy.¹⁵⁶

Management of epilepsy in general practice

In acute epilepsy, diazepam, lorazepam and midazolam are the drugs most commonly used to control prolonged seizures.¹⁵⁵

In chronic epilepsy, the use of benzodiazepines is now rare. They have limited usefulness in long-term management due to the rapid development of tolerance to anticonvulsant effects and the associated side effects (eg sedation and psychomotor slowing).¹⁵⁵

Benzodiazepines are considered adjunct to standard antiepileptic drugs, and used when these have failed to achieve acceptable control.¹⁵⁵

Benzodiazepines for seizure prophylaxis should only be used on advice from a specialist neurologist or paediatrician.

Prescription of benzodiazepines for epilepsy within Australia requires an authority prescription. Take care with monitoring and detecting problematic use.