Antidepressants
| Type | Recommendation |
|---|---|
| When to deprescribe | |
| CBR |
We suggest deprescribing be offered to older people taking antidepressants for major depressive disorder:
|
| Ongoing treatment | |
| CBR |
If deprescribing is unsuccessful despite multiple attempts, taking into account the possibility of withdrawal effects rather than recurrence of symptoms, we suggest maintaining the lowest effective dose; however, we suggest reassessing the need for long-term therapy periodically. |
| How to deprescribe | |
| CBR |
We suggest individualising the tapering schedule and adjusting it according to the individual's response. In general, we suggest reducing the dose by 25% to 50% every one to four weeks (taking into consideration the half-life of the antidepressant), ensuring the absence of physical or neuropsychiatric withdrawal symptoms before initiating further tapering. Once half the lowest standard dose formulation is reached for another one to four weeks, we suggest ceasing completely if no sign of reoccurrence of symptoms. However, smaller dose reductions may be appropriate or preferred by some individuals, particularly as lower doses are approached. |
| GPS |
Healthcare providers should consider and offer adequate non-pharmacological management options (e.g. psychological interventions for psychiatric disorders such as cognitive-behavioural therapy) to individuals and their families or carers as appropriate (ungraded good practice statement). |
| Monitoring | |
| CBR |
We suggest closely monitoring for worsening neuropsychiatric symptoms (e.g. increased anxiety, agitation, depressive symptoms) and cognition which could be short-lived or protracted, severe or mild, in addition to monitoring changes in psychological or physical health status, and quality of life every one to two weeks following each dose adjustment until at least four weeks after the medicine is fully ceased if practical (recognising the possibility of withdrawal effects rather than recurrence of symptoms). After this initial period, we suggest monthly monitoring for at least three months, followed by monitoring every six months thereafter. However, this should be tailored based on individual factors such as their preferences, responses and tolerance to deprescribing. If in-person visits are impractical, we suggest advising people to report symptoms as needed. |
CBR, consensus-based recommendation; GPS, good practice statement
Clinical depression is common in older people and should not be mistaken as a normal part of ageing. Poor mental health is particularly prevalent among those in residential aged care, where 87% have at least one mental health or behavioural condition, and 49% are diagnosed with mood disorders, including depression [50]. Psychotherapies such as Cognitive Behavioural Therapy (CBT) are effective in reducing depressive symptoms in older adults [694].
When antidepressant therapy is indicated, SSRIs and SNRIs are generally preferred in older people due to their lower risk of adverse effects and relative safety in overdose compared to tricyclic antidepressants (TCAs) [695]. However, all antidepressants have potential risks, including hyponatremia, falls, and gastrointestinal bleeding [177]. The optimal duration of therapy and the criteria for discontinuation remain unclear in the literature [696].
Refer to the narrative evidence summary, the GRADE Summary of Findings table in the guidelines, and the Technical Report for a complete presentation of the deprescribing evidence based on the GRADE framework (including other factors considered in developing the recommendations).
A systematic review and meta-analysis examined the risk of relapse after discontinuing antidepressants in individuals with major depressive disorder who had achieved remission [697]. The participants in the meta-analysis had a mean age of 43 years (n = 8,890), younger than the target population for this guideline. Those who continued antidepressants for six months had a significantly lower relapse rate compared to those who switched to a placebo. Similarly, a pooled analysis of 45 RCTs found that relapse risk was minimal when antidepressants were continued for at least four to six months after stabilisation [698]. These findings support maintaining antidepressant therapy for at least six months following remission, aligning with the 2023 Canadian Network for Mood and Anxiety Treatments (CANMAT) guidelines, which recommend continuing antidepressants for six to 12 months post-remission to prevent recurrence [699].
In people who have achieved symptomatic remission or clinical stability for six to 12 months with uninterrupted treatment after appropriate assessment, offering to attempt deprescribing may be appropriate. Besides, in people where no benefit has been derived from antidepressants, deprescribing should be considered given the unfavourable benefit-risk profile.
Long-term antidepressant use should be regularly reassessed, particularly when initiated for behavioural and psychological symptoms of dementia (BPSD). Ongoing monitoring should consider the indication, effectiveness, tolerability, and iatrogenic risks [50].
When deprescribing antidepressants, the tapering plan should be individualised to minimise withdrawal symptoms and the risk of relapse. A general approach involves reducing the dose by approximately 25% every one to four weeks, with a slower taper for final dose reductions or if discontinuation symptoms arise [627]. Slower tapering could be an appropriate suggestion, especially for people at risk of withdrawal symptoms while tapering (e.g. those who have previously experienced withdrawal symptoms when they have missed a dose), if problematic discontinuation symptoms occur, or when approaching the final dose reduction [627, 700]. In practice, the tapering approach will also depend on the half-life of the antidepressant and its metabolism. Antidepressants with long half-life (e.g. fluoxetine) can be discontinued by staggering the days of use, whereas those with shorter half-life may require more judicious daily dose reduction [701].
The 2024 Maudsley deprescribing guidelines recommend a non-linear, hyperbolic tapering strategy for psychotropic medications, including antidepressants, to minimise the risk of relapse and withdrawal symptoms [43]. This approach involves progressively smaller reductions at lower doses and is based on recent findings of the relationship between the dose and the serotonin receptor occupancy [702]. For example, a suggested tapering schedule for citalopram (20 mg, 9.1 mg, 5.4 mg, 3.4 mg, 2.3 mg, 1.5 mg, 0.8 mg, 0.4 mg) approximates 10% reductions in serotonin receptor occupancy with each citalopram dose reduction. The hyperbolic tapering strategy offers a gradual approach that may be preferred by some individuals. Implementing a slower or hyperbolic tapering regimen may require access to liquid formulations or compounded preparations to achieve the small, precise dose reductions necessary. While these considerations are important for individualised care and may reflect patient preferences, they may also present practical challenges in some settings. Notably, no studies were identified that directly evaluated the impact of these formulation requirements on the feasibility or uptake of antidepressant tapering strategies.
At the time of this guideline's development, the reporting of the RELEASE (REdressing Long-tErm Antidepressant uSE in general practice) trial is ongoing [703]. This study evaluates the effectiveness of multi-strategy interventions for the safe discontinuation of long-term antidepressants. It is likely that future guideline recommendations will evolve as new evidence emerges.
From the systematic review and meta-analysis, we identified four studies related to SSRI deprescribing (two RCTs, two before-and-after studies), one before-and-after study related to deprescribing nortriptyline/phenelzine (with or without adjunctive lithium) and two studies related to deprescribing lithium augmentation for depression (one RCT, one case-control study) [632, 704-709].
Overall, the current evidence for antidepressant deprescribing is derived from studies of low and very low certainty. While deprescribing was shown to lead to a significant deterioration in neuropsychiatric symptoms in an RCT, this outcome is of very low certainty. The evidence to date is derived from studies of small sample sizes and generally have serious methodological limitations. The studies targeting SSRIs were all conducted in nursing home settings and had a very short follow-up duration (6-12 months). It may be possible that severe and persistent withdrawal syndromes, if occurred, were not captured. It was uncertain whether the reported outcomes were applicable to other settings. There was also a lack of clear differentiation between withdrawal effects and relapse of the initial condition in many studies. The evidence at this stage remains insufficient to inform evidence-based recommendations.
If antidepressants are considered appropriate to deprescribe, closely monitoring for any worsening of neuropsychiatric symptoms (e.g. increased anxiety, agitation, or depressive symptoms), as well as changes in psychological or physical health, quality of life, and cognitive function may be appropriate. It is also important to recognise that withdrawal effects from some antidepressants can be severe and, in some cases, protracted [43]. Withdrawal effects should be distinguished from a recurrence of the underlying depressive symptoms to minimise unnecessary reinitiation of antidepressants.
Key study characteristics and results
A narrative summary of each study is provided below, highlighting key characteristics and main findings.
Bergh 2012 conducted a double-blind RCT that randomised 128 nursing home residents to discontinuation (n=63) or continuation (n=65) [704]. Residents were included if they had been taking either escitalopram, citalopram, sertraline, or paroxetine for at least three months, or had Alzheimer's disease, dementia or vascular dementia, and neuropsychiatric symptoms. They were excluded if they had a history of a depressive disorder or schizophrenia. Among the 81 participants who completed the 25-week follow-up (35 intervention, 46 control), discontinuation of antidepressants was not associated with a significant deterioration in the Cornell scale of depression in dementia (MD 1.61, 95% CI -0.39, 3.61). There was no significant difference between the two groups in terms of mortality, body weight, number of falls, self-care ability, physical function measured using the Unified Parkinson's Disease Rating Scale, cognition measured using the Severe Impairment Battery, quality of life measured using the Qualify of Life in Alzheimer's Disease scale, the amount of rescue medicine used (measured in oxazepam mg/day over 21 days), or the total number of psychotropic medicines used by participants. However, neuropsychiatric symptoms measured using NPI were more pronounced in the intervention group than in the control group (MD 7.80, 95% CI 1.10 to 14.50).
Ulfvarson 2003 randomised 70 nursing home residents to discontinuation (n=35) or continuation (n=35) [709]. Residents were included if they had been taking an SSRI (sertraline, citalopram) for at least six months without any documented indication or symptoms of depression or anxiety disorder. Additionally, they had to score 12 or less on the Montgomery-Asberg Depression Rating Scale score to be included. Residents with dementia or a history of depression were excluded. Among the 52 participants who completed six months of follow-up, there were no significant differences between the two groups in terms of mortality, depressive symptoms measured using the Montgomery-Asberg Depression Rating Scale (MADRS), functioning measured using the global assessment of functioning, symptoms of depression and side effects to SSRIs (on a 0-100 point scale, where a higher score indicates greater side effects of depression or SSRIs), or symptoms of common side effects to sertraline and citalopram (on a 0-52 point scale, where a higher score indicates worse symptoms). Quality of life measured using the Health Index was significantly lower in the intervention group than in the control group (MD 1.72, 95% CI 0.11 to 3.33).
Lindström 2007 conducted a before-and-after study that included 119 nursing home residents who had been taking SSRIs for at least 12 months [708]. Residents were excluded if they were using SSRIs for indications other than depression there was a long-term indication, current depressive symptoms, or they had two or more episodes of depression in the past two years. Deprescribing was successful in 63/119 (53%) participants and was reported to be more likely in residents with a low to moderate MADRS score (0-19) prior to deprescribing.
Bergh and Engedal 2008 conducted a before-and-after study that included 23 nursing home residents with Alzheimer's disease or vascular dementia who had been taking antidepressants or antipsychotics for three months or longer [632]. Deprescribing was implemented for antidepressants (n=11) or antipsychotics (n=12). Participants were excluded if they had a severe psychiatric disorder or diabetes mellitus. At 24 weeks, discontinuation of antidepressants was associated with non-statistically significant improvement in BPSD measured using the Neuropsychiatric Inventory (29.2 ± 20.2 to 17.3 ± 21.4), depression measured using Cornell's Depression Scale (6.9 ± 4.5 to 3.3 ± 3.4) and movement disorders measured using the Unified Parkinson Disease Rating Scale (6.4 ± 4.2 to 4.5 ± 3.4). However, cognition, measured using the Severe Impairment Battery, appeared to deteriorate after antidepressant withdrawal (50.1 ± 22.5 to 28.0 ± 20.3).
Flint 1999 conducted a before-and-after study that included 21 patients who were taking nortriptyline (with or without adjunctive lithium) or phenelzine and had not experienced a relapse or depression recurrence for the past two years [706]. Patients with a concurrent axis I diagnosis, history of schizophrenia, schizoaffective disorder, paranoid disorder, dementia, or any neurological disorder were excluded. Deprescribing was successful in 9/21 (43%) participants. Of the 12 participants who had depression recurrence, 11 (92%) restarted their antidepressants. Reintroduction of antidepressants alone led to improvements for 10/11 participants (91%) who responded in a mean (SD) of 4.5 ± 1.8 weeks.
The following two studies reported deprescribing lithium augmentation therapy for major depressive episodes (with or without psychotic features).
Fahy 2001 reported a case-control study that included 21 patients who were on lithium augmentation but subsequently discontinued [705]. Antidepressant medicines were continued during the observation period. The study reported that 11/21 (52%) had a recurrence of depression, of whom 9/11 (82%) had responded to the reintroduction of lithium and 2/11 (18%) responded to another antidepressant.
Hardy 1997 randomised 12 older people with unipolar depression to receive continued lithium augmentation (n=6) or matching placebo (n=6) [707]. Participants were included if they had not experienced depressive symptoms for at least 12 months while on a stable dose of lithium augmentation, scored < 20 on Geriatric Depression Rating Scale and scored < 15 on MADRS. Participants were excluded if they had dementia or suicidal tendencies. There was no significant difference between the two groups in terms of depression recurrence (OR 1.00, 95% CI 0.09, 11.03) and thyroid stimulating hormone levels (MD 0.56, 95% CI -0.08, 1.20). Among the six participants who received a placebo, two (33%) reported a recurrence of depression at seven and 92 weeks respectively, without any apparent changes in life stresses and were relatively resistant to reinstitution of lithium augmentation therapy. Comparatively, 2/6 participants (33%) who continued lithium augmentation also reported a recurrence of depression immediately after a stressful life event at 46 and 61 weeks respectively. Control group participants who continued to receive lithium over the two years appeared to have a significantly higher serum creatinine level compared to participants who received a placebo (MD 13.30, 95% CI 0.47, 26.13).
Various methods were used for deprescribing in the included studies and tapering was the most common method. While there was no direct evidence that any particular method was associated with the greatest benefits and harms, dose tapering is likely more acceptable than abrupt cessation and helpful in determining the lowest effective dose for some patients requiring dose reduction rather than complete cessation.
In the two RCTs targeting SSRIs, the dose was halved for a few days before complete withdrawal (n=70, very low certainty evidence) [709] whereas the method was not described in the other RCT (n=128, very low certainty evidence) [704]. In the two before-and-after studies targeting SSRIs, the dose was tapered gradually over six to eight weeks before complete withdrawal (n=119, did not report critical/important outcomes) [708] whereas in another study dose was tapered over one week (n=11, very low certainty evidence) [632]. Nortriptyline and phenelzine were tapered over eight weeks (n=21, very low certainty evidence) [706].
In the RCT targeting lithium augmentation, the dose was reduced by 150 mg daily each week until completely replaced with a matching placebo (n=12, low certainty evidence) [707]. In the case-control study, lithium augmentation was tapered gradually over a period of two to 12 weeks (n=21, very low certainty evidence) [705].