Quality Prescribing for Antidepressants: A Guide for Improvement 2024-2027

3. Which groups of people should be targeted for review?

This may vary depending on practice populations and prescriber preferences, however, we would advise that the following groups should be considered for review. The Scottish Therapeutics Utility (STU) is available in all general practices in Scotland to allow case finding and identification of the higher risk groups below.

List 1: Potential groups of people that receive antidepressants and may benefit from being prioritised for a regular, proactive medicines review

• People receiving the same antidepressant continuously, long-term (≥2 years)
• Older adults (≥65 years) and/or frail adults
  • Receiving a SSRI plus antiplatelet/NSAIDs/DOAC/warfarin without gastro-protection, due to the increased risk of gastric bleed with this combination.
  • Receiving >1 defined daily dose per day of citalopram (>20mg per day) or escitalopram (>10mg per day) due to QTc prolongation
  • Diagnosed with dementia,
    • Antidepressants may provide limited effects for depression
    • Receiving antidepressants for behavioural and psychological symptoms of dementia
  • Antidepressant plus long-term benzodiazepines (≥8 weeks) and/or z-drugs (B-Z)
    • B-Z are associated with an increased risk of depressive symptoms, paradoxically worsen anxiety, and reduce the effectiveness of psychological treatments. Review reducing B-Z use prior to assessing antidepressant need where appropriate.
    • B-Z prescribing and use is associated with use of higher doses of SSRIs
  • Combination antidepressant treatment e.g. mirtazapine with SSRI/SNRI. There is a lack of evidence demonstrating the efficacy of this combination to treat depression. People initiated on combinations by psychiatry should be reviewed by specialist services.
  • Higher risk of QTc prolongation combinations:
    • Citalopram or escitalopram or TCAs plus: methadone, antipsychotics, quinine, some antinausea medicines, etc
• People receiving high dose SSRIs for the treatment of depression
• People receiving low dose mirtazapine or trazodone for insomnia or subtherapeutic doses of mirtazapine (<30mg per day) for the treatment of depression

Receiving TCAs and other anticholinergics – reduce the anticholinergic drug burden, to reduce the risk of falls and confusion

People receiving combinations:

Note: DOAC: direct oral anticoagulant. NSAID: non-steroidal anti-inflammatory drug. SSRI: selective serotonin re-uptake inhibitor. TCA: Tricyclic antidepressant QTc prolongation see Section 1 for more information.

People receiving the same antidepressant long-term, for two or more years

As already acknowledged, this is a growing population.^[2],^[3] This group is less likely to have their antidepressant reviewed than people who have recently been initiated on treatment or are receiving a shorter course.^[4],^[5] For some they may be ‘lost in the system’ and inadvertently continue treatment that is no longer needed.^[14],^[22] For others their antidepressant treatment may be ineffective or causing adverse effects. Therefore, it is necessary to proactively review individuals’ progress, and while some guidelines provide clear advice regarding the frequency of review,^[36],^{[37], [89]} others remain vague.^[20],^[50],^[90],^[91] For those with chronic illness, where long-term treatment is considered necessary, annual reviews may provide an optimal method e.g. assessing the need for TCA/duloxetine treatment of neuropathic pain from diabetes.

For those under the care of specialist secondary care services (e.g. Pain Clinics, Alcohol and Drug Rehabilitation Services, Community Mental Health, Learning Difficulties, etc.), it is important that the specialist services update primary care prescribers about the appropriateness of continuing treatment. In some cases, this will facilitate the availability of prescribing information on the patient’s Emergency Care Summary.

Reviewing people receiving the same antidepressant for two or more years can result in one in four people having a change in treatment and some reducing and/or stopping their antidepressants.^[17]

Frail and older adults: avoidable adverse drug events, dementia and polypharmacy and high-risk combinations

All antidepressants are associated with increasing risk of falls in older adults.^[43] SSRIs are also associated with an increased risk of gastrointestinal bleed, which is further increased when they are used in combination with antiplatelets, non-steroidal anti-inflammatory drugs (NSAIDs), direct oral anticoagulants (DOACs), and/or warfarin.^[92],^{[93], [94]},^[95] Reviewing people in this group can ensure appropriate use and minimise avoidable adverse drug events and harms. For some individuals initiation of appropriate gastro-protection such as proton pump inhibitors may be advised in local and national guidelines.^[20],^[77]

QTc prolongation can occur with a range of antidepressants and other medicines. ^[96],^[97] However, higher doses of citalopram and escitalopram are associated with an increased risk of QTc prolongation. Therefore, all older adults (>65 years) prescribed more than citalopram 20mg or escitalopram 10mg daily should be reviewed and considered for dose reduction and/or cessation where appropriate. ^[98] Where it is considered clinically appropriate, or people refuse to reduce their citalopram or escitalopram dose, discuss and document the risk of harm and arrange regular cardiac QTc monitoring as per List 2. Older and/or frail adults are at higher risk of QTc prolongation which is associated with ventricular tachycardia and sudden cardiac death.^[51] In part this due to ageing but can be exacerbated by comorbidities and multiple medicines e.g. antibiotics, cardiac, diuretic, psychotropics, respiratory, etc.^[6] Therefore, proactively reviewing polypharmacy will also help to reduce the risk of QTc prolongation and sudden death.

A list of medications known to prolong the QT interval^[99] can be found on Stockley’s Drug Interactions and the Credible Meds website.

List 2: Monitoring criteria for citalopram and escitalopram^[98]

• People with cardiac disease, consider an ECG review before starting treatment with citalopram and escitalopram.
• Electrolyte disturbances (e.g. hypokalaemia and hypomagnesaemia) should be corrected before treatment with citalopram and escitalopram. Monitoring of serum magnesium is advised, particularly in older adults, who may be taking diuretics or proton pump inhibitors.
• If cardiovascular symptoms, such as palpitations, vertigo, syncope, or seizures develop during treatment, cardiac evaluation including an ECG should be undertaken to exclude a possible malignant cardiac arrhythmia.
• If QTc interval is >500 milliseconds, treatment should be withdrawn gradually.
• If QTc interval duration is between 480 milliseconds and 500 milliseconds, the balance of benefits and risks of continued treatment should be carefully considered, alongside options for dose reduction or gradual withdrawal.

Dementia

For dementia, antidepressants demonstrate limited benefits in treating depression.^[18],^[100],^[101],^[102] However, for some individuals they may reduce depressive symptoms and improve general functioning.^[103] There are relatively few studies of antidepressants for the treatment of agitation and psychosis in dementia, however sertraline, citalopram and trazodone have been used and are associated with modest reductions in symptoms of agitation and psychosis.^[104],^[105] Antidepressants also demonstrate mixed and limited effects for the treatment of behavioural and psychological symptoms of dementia,^[18] therefore, routine reviews may help to appropriately minimise doses and optimise non-pharmacological management.

Polypharmacy and anticholinergic effects

TCAs have strong to very strong potential for anticholinergic risk.^[106] Consider TCA use in relation to other prescribed and non-prescribed medicines with anticholinergic effects. These include antihistamines, anti-Parkinson’s medicines, urinary antispasmodics and some antinausea medicines, etc. [see Polypharmacy Guidance for more information].^[9] TCA anticholinergic effects include an increased risk of dry mouth, blurred vision, cognitive dysfunction, urinary retention and falls, etc.^[31]

People receiving antidepressants in combination with other psychotropics

Benzodiazepines and z-drugs (B-Z) with antidepressants

All B-Z use can lead to long-term regular use, sometimes lasting for years. ^[107] This is contrary to good practice guidance^[36],^[37] and the terms of their licence^[31],^[69] therefore, consider reviewing appropriateness where:

• B-Z are initiated to treat anxiety or insomnia prior to starting antidepressant therapy, or to treat agitation, anxiety or insomnia symptoms associated with starting an SSRI.^[20],^[36] B-Z only demonstrate marginal benefits for short-term relief of insomnia and some anxiety disorders.
• B-Z are initiated to treat avoidable adverse drug effects caused by higher SSRIs doses (e.g. insomnia, agitation),^[39],^[40],^[42] or for signs and symptoms of poorly controlled depression, anxiety or back pain.
• B-Z use is known to worsen depressive symptoms, cause cognitive dysfunction and other avoidable adverse effects, and reduce the efficacy of some psychological therapies.^[107],^[108],^[109],^[110] This should be considered one of the priority groups for review.
• Long-term B-Z have been prescribed, to review and gradually withdraw using an agreed structured and planned reduction schedule. A small minority of individuals may require longer-term B-Z treatment with regular review to optimise care and minimise street B-Z use, see the Benzodiazepine and Z-drugs Quality Prescribing Guidance for more detail.
• Accompanying this guide are a suite of National Therapeutic Indicators allowing identification of variation in prescribing at NHS board or GP practice level with accompanying case finding STU searches to allow identification of individuals at risk of harm from within general practice. For example, those on antidepressant medication in combination with other psychotropic medication.

Combining antidepressants for depression

Combining antidepressants for depression is not recommended. Non-specialist psychiatry prescribers should not initiate such combinations, unless on the advice of specialist services.

• Consultant psychiatrists may initiate combined therapy for their regular patient(s), e.g. for treatment of treatment resistant depression. The quality of evidence supporting combination antidepressant treatment is poor. For example, SSRIs or serotonin and noradrenaline re-uptake inhibitors (SNRIs) plus mirtazapine (30-45mg) demonstrate marginal benefits in observational studies to no difference in randomised placebo-controlled studies.^[20],^[111],^[112]
• Antidepressant augmentation with an antipsychotic or lithium therapy (as per guidance from the Directorate of the Chief Medical Officer)^[113] can be more effective; providing clearer benefits, but these are not without risks and the requirement of extra monitoring which is often lacking.^[20],^[114],^[115] Monitoring for antipsychotic therapy may include regular cardiometabolic risk assessment, including HbA1c, lipids, weight, blood pressure and lifestyle advice.^[116]

Low dose mirtazapine added to SSRI/SNRIs therapy:

• This combination to reduce antidepressant induced sexual dysfunction is supported by limited evidence of efficacy.^[117],^[118] Clinicians should assess the efficacy of such strategies and record this as the mirtazapine indication where appropriate.
• This combination has also been used for short-term sedating antihistamine effects (also seen with trazodone). In part, these additional antidepressants are possibly being used to treat SSRI/SNRI induced insomnia, agitation etc., and as an alternative to B-Z. However, this combination is of questionable benefit, and it may be more appropriate to reduce the dose of the SSRI/SNRI to minimise potential adverse effects and drug-related harms, rather than adding additional psychotropics, especially as tolerance can develop quickly to mirtazapine and trazodone sedating effects.^{[84], [119].} This avoids the prescribing cascade.

Specialists should be conscious of their prescribing actions potentially influencing prescribing habits in general practitioners.5 Specialists should consider providing clear rationale and therapeutic goals, alongside withdrawal strategies when initiating psychotropic combinations as outlined above.

Neuropathic pain plus depression/anxiety:

• At times prescribers switch from a TCA and an effective antidepressant (for depression for example) to a TCA or duloxetine to treat both pain and mood. In these cases the TCA or duloxetine may only be effective for one condition, and not the other, or individuals may not tolerate higher doses of TCAs or duloxetine which can be more effective for treating depression.^[20],^[39],^[53],^[54] For non-neuropathic pain prescribers should also consider avoiding concomitant antidepressant and tramadol use, due to the risk of serotonin syndrome.^[94]
• Two antidepressants (e.g. TCA for neuropathic pain and an antidepressant for depression such as a SSRI) may be appropriate to reduce symptoms (not two TCAs). However, use should be reviewed regularly, considering interactions, adverse and synergistic effects, e.g. sedating effects, TCA dose related QTc prolongation (greater risk of prolongation with higher doses).^[96]

High dose SSRIs for the treatment of depression

Standard SSRI doses provide the optimal balance between efficacy and minimising adverse effects and harms. This is due to their flat dose response curve for the treatment of depression, meaning that ‘20’s plenty and 50’s enough’ to provide the full antidepressant effect. (Standard doses provide optimal serotonin transporter occupancy.)^[120] Previously prescribers have been heavily criticised for prescribing subtherapeutic licensed doses of TCAs for the treatment of depression. (Licensed dose is the approved dose or dose range that medicines are licensed to be prescribed at by the regulatory authority – Medicines and Healthcare products Regulatory Agency (MHRA) – for the condition that they approved to treat, e.g. sertraline for depression treatment has a dose range of 50mg to 200mg per day.^[121],^[122] While TCAs and SNRIs demonstrate dose response effects, with larger doses being more effective for depression treatment,^[39,40,53] higher than standard daily doses of SSRIs (20mg citalopram/fluoxetine/paroxetine, 50mg sertraline or 10mg escitalopram) do not provide better response rates, not even for poor or partial responders.^[39-42],^[44]

While a range of campaigns and guidelines for depression have heavily promoted the message ‘to increase the dose’ of antidepressants for poor and non-responders,^[36],^[123],^[124],^[125] a minority of guidelines have highlighted the differences in response and efficacy between SSRIs and other antidepressants.^[20],^[126] Individuals and prescribers may have expectations that higher doses are more effective for routine treatment of depression,^[5],^[11] in part this may be due to guidelines or training, as well as individual or societal expectations and beliefs regarding medicines.^[29],^[127],^[128] For a small minority of individuals, where a trial of higher dose SSRI is considered appropriate, follow up review within two to four weeks should be arranged to assess response to treatment and address any adverse effects. It is known that early improvement predicts a stable response.^[73],^[129] For anxiety disorders SSRI dose response effects are mixed. GAD guidelines indicate no clear indication of a dose response relationship, whereas obsessive compulsive disorder guidelines indicate that higher SSRI doses can be more efficacious.^[37],^[69],^[89]

In 2011, in the UK, the MHRA issued advice regarding new maximum daily dose restrictions, contraindications, and warnings for citalopram and escitalopram use.^[98] Health boards issued local advice on reviewing citalopram/escitalopram doses: reviewing, reducing doses and if necessary switching to an alternative antidepressant where appropriate.^[130] Anecdotally from prescribers and general practice feedback, of those individuals that were reviewed: some stopped, some required a switch to an alternative antidepressant; while the vast majority that were required to continue treatment were continued on lower doses, as per MHRA advice, without worsening of their depressive symptoms.

Low dose mirtazapine and trazodone for sedation

Low daily doses of mirtazapine (15mg) or trazodone (100mg or less), are commonly used to treat insomnia and anxiety symptoms due to their sedating antihistamine effects.^[84],^[119] However, people commonly develop tolerance (usually within a couple of weeks) to such sedating effects and therefore longer-term use – as with B-Z drugs – may be inappropriate.

Antidepressants demonstrate mixed and limited effects for the treatment of behavioural and psychological symptoms of dementia,^[18] therefore, routine reviews may help to appropriately minimise doses and optimise non-pharmacological management.

The routine use of low dose mirtazapine for the treatment of depression should be reviewed, as the majority of clinical trials demonstrated efficacy with doses ≥30mg daily.^[26],^[40],^[131] Therefore, if mirtazapine is considered as appropriate and necessary then increasing to 30mg per day within the first week of treatment with an agreed plan for follow up within two to four weeks to assess efficacy and tolerance may help to provide an optimal response.

Pharmacogenomic testing and antidepressant therapy

Pharmacogenomic testing has the potential to improve medicines optimisation by providing information on the individual’s response to treatment, improving effectiveness and safety of treatment by reducing the risk of adverse drug reactions (ADRs). People have on average four gene variants that influence medication effectiveness and risk of ADRs, with 99% of individuals having at least one gene variant that affects drug metabolism. The use of pharmacogenetics to guide treatment choices could reduce the likelihood of medication related harm and improve patient outcomes from safer personalised medication regimes. Some studies have shown that this approach could be more cost effective, and that pharmacogenomics can be used to support antidepressant prescribing,^[132] however more evidence is needed together with a tested framework for implementation.