In 2018, Public Health England reported that one in five older people living in the community and two in five older people living in care homes are affected by depression. These symptoms are associated with reduced quality of life and high morbidity, and also with increased mortality through suicide and self-neglect. A 2014 study showed that major depression was associated with a 43% increase in the risk of non-suicide-related mortality in adults over the age of 50. Depression across all age groups has a detrimental impact on recovery from surgery and in older people, increases the risk of coronary heart disease and stroke., There is a gathering body of evidence that depression is a risk factor for dementia and, that in people with mild cognitive impairment, the presence of depression may increase the risk of progression to dementia.
In 2018, Public Health England reported that one in five older people living in the community and two in five older people living in care homes are affected by depression.1 These symptoms are associated with reduced quality of life and high morbidity,2 and also with increased mortality through suicide and self-neglect.3 A 2014 study showed that major depression was associated with a 43% increase in the risk of non-suicide-related mortality in adults over the age of 50.4 Depression across all age groups has a detrimental impact on recovery from surgery, and, in older people increases the risk of coronary heart disease and stroke.5, 6 There is a gathering body of evidence that depression is a risk factor for dementia and that in people with mild cognitive impairment the presence of depression may increase the risk of progression to dementia.7
Depressive disorders presenting in older people have many clinical similarities to depression in younger adults; however, biological changes related to ageing may necessitate a different approach to treatment. In this chapter, we present an evidence-based review of treatment for late-life depression, focussing on pharmacological approaches including monotherapy, combination and augmentation strategies. Selective serotonin reuptake inhibitors (SSRIs) including sertraline and citalopram are well tolerated with the advantage of a favourable side-effect profile and are good options for first-line treatment. Second-line treatment options include combination therapy with a second antidepressant, or treatment augmentation with anti-psychotic medication or lithium. We also consider evidence for non-pharmacological treatment strategies including psychological therapy and neuro-stimulation. Finally, we summarize evidence for treatment of depression in patients with dementia.
Diagnosis of late-life depression is based on standard DSM-5 or ICD-11 criteria for mood disorders.8, 9 There may be subtle differences in clinical presentation, with higher rates of somatic symptoms, apathy, psychomotor retardation or agitation, fatigue and executive dysfunction compared with younger age groups,10 but these symptoms are not sufficiently different or reliable to classify this as a distinct disorder.
The aetiology of late-life depression is multifactorial, and includes biological, psychological and social factors (see Figure 18.1). These aetiological factors can lead to relapse into a depression in those with an existing vulnerability, or to the first onset of a depressive disorder in later life. Since many of the aetiological factors are age-related, there has been (and continues to be) a misconception that depression in older people is inevitable given the frequency of physical ill health, bereavement and loss of role in this age group. This prejudice may reduce the likelihood of help-seeking behaviour by patients and of the offer of treatment by health professionals and could be one reason why late-life depression remains an under-recognized and undertreated disorder. This may be a particularly prevalent prejudice affecting those living in nursing homes.11 Given the significant adverse impact on patients, their families and carers, late-life depression warrants greater efforts at recognition and effective, evidence-based treatment.
Prescribing for Older Adults
Pharmacotherapy can be used effectively in late-life depression.12 However, treatment requires a distinct approach because of age-related physiological changes, which alter pharmacokinetics and pharmacodynamics. Drug distribution within the body changes because of increased body fat and reduced volume of water. Hydrophobic compounds (e.g. antidepressants and antipsychotics) are therefore more readily distributed, whereas there is reduced distribution of hydrophilic compounds. Renal excretion is reduced, so there may be increased concentrations of active compounds that are renally excreted. These changes mean that lower doses, slow titration and careful monitoring are required.13, 14 The increased prevalence of medical co-morbidities in this age group means that patients are more vulnerable to medication side effects. Polypharmacy is a risk, and care is needed to minimize the effects of drug interactions.15 These considerations should not discourage the use of pharmacological treatment of late-life depression, but do necessitate a careful review of the person’s past medical history before drug regimens are changed.
This chapter provides an overview of treatments for late-life depression (pharmacological, psychological and neurostimulation), based on a search of electronic databases including Medline, Embase, PsychINFO and PubMed for the terms [treatment or therapy] and [depression or depressive] and [old age or older or late-life or geriatric] between 1990 and January 2019. This is intended to summarize important and recent research findings in a format that is clinically relevant.
SSRIs are the most common first-line treatments for late-life depression, similar to treatment for younger patients. Sertraline has the best evidence for efficacy and tolerability and should be considered first-line for late-life depression.18 There is also good evidence for escitalopram and citalopram in those over 65 years old; these have a low propensity for drug interactions as they have minimal impact on the cytochrome P450 system.12, 17 All three have the advantage of a favourable side-effect profile and can be used in patients with a history of cognitive impairment and stroke.18
A large cross-sectional study has confirmed a modest dose-dependent QT interval increase with citalopram (10 mg to 20 mg, mean QTc increase 7.8 ms) when compared to other SSRIs.19 The maximum recommended dose for older adults has now been reduced and citalopram should not be prescribed in conjunction with other agents which prolong the QT interval, such as antipsychotics and certain antibiotics (e.g. erthyromycin). SSRIs also carry a risk of hyponatremia which increases with age. Female gender, previous hyponatraemia or concurrent diuretic or non-steroidal anti-inflammatory drug (NSAID) prescription have all been demonstrated to confer additional risk and monitoring of urea and electrolytes should be considered in high-risk individuals.20 SSRIs should be avoided in those with a history of gastrointestinal bleeding, especially if they are prescribed other medications which may precipitate this.
Some studies have cast doubt on the efficacy of SSRIs in late-life depression.21 Where SSRIs are contraindicated, or poorly tolerated, other classes of antidepressants can be used as first-line treatment.
Mirtazapine is an alternative to SSRIs for first-line treatment in older adults.12, 22 It has few cardiac side effects and a weaker association with hyponatraemia; in patients with insomnia and anorexia its side effects may be exploited for therapeutic benefit. It appears to have a faster onset of action than SSRIs, with beneficial effects found at two weeks.23
Venlafaxine has been proposed as a useful second-line antidepressant.12 An open-label study demonstrated the efficacy of venlafaxine in older patients with atypical subtypes of depression including mood reactivity, hypersomnia and increased appetite.23 Venlafaxine may, however, exacerbate hypertension and therefore close monitoring of this is needed at treatment initiation.
Both vortioxetine and agomelatine have had positive placebo-controlled trials in people aged over 65 years.24, 25 Vortioxetine, a serotonin transporter blocker, does not have a significant impact on QTc and has been found to improve cognitive tests of processing speed, verbal learning and memory in people over 65 years when compared to placebo. Nausea and vomiting can, however, be a prominent side effect in clinical practice. Agomelatine is very well tolerated but requires monitoring of liver function. A recently updated network analysis ranked agomelatine among the most efficacious antidepressants and it may be particularly useful in individuals who are prone to side effects.26
Tricyclic antidepressants are also effective treatments in older age groups. Nortriptyline shows comparable efficacy and tolerability when compared with sertraline.27 Adverse effects were similar in both groups, but tricyclic antidepressants should be used with caution owing to the risk of cardiac side effects, toxicity in overdose and because their anti-muscarinic effects are more likely to exacerbate cognitive impairment.
A meta-analysis of late-life depression found studies with a longer trial period (10–12 weeks) were significantly more likely to demonstrate a response, suggesting older people may take longer to respond to antidepressants – at least six weeks.13 If there is only a modest improvement at this stage and the antidepressant is tolerated, the dose should be increased.30 Where first-line monotherapy has been ineffective, switching to a second-line agent (in the same, or a different class) is a useful strategy, particularly for those who received minimal benefit, or could not tolerate the first-choice antidepressant.30 If there has only been a partial response after the dose is optimized to the maximum specified or tolerated dose, augmentation should be considered.
Although common in clinical practice, there is little evidence that combining different antidepressants is effective in older people.31 Augmentation with other agents may be effective, albeit with slower and lower recovery rates; medical co-morbidity and symptoms of anxiety are associated with a poorer prognosis.32 As in younger populations, lithium has the best evidence base as an augmenting agent with a systematic review reporting an overall response rate of 42% (95% confidence interval [CI] 21–65%).33 For older adults, this should be accompanied by careful monitoring of renal and thyroid function, using lower doses than in younger patients.
Augmentation of antidepressant treatment with antipsychotic medication is useful for those with severe depression, particularly where there are psychotic symptoms, providing there has been careful consideration of the risks and benefits.34 For psychotic depression in older adults, treatment with olanzapine and sertraline has been associated with significantly higher remission rates than treatment with olanzapine alone.35 The use of olanzapine is associated with increased cholesterol and triglyceride concentrations; associated weight gain, however, may be less significant in older adults compared to younger adults.35 In a randomized, placebo-controlled study of older patients with resistant depression, augmentation of citalopram with risperidone led to symptom resolution in a substantial number of patients, and a delay in time to relapse (56% in the risperidone group and 65% in the placebo group relapsed). 36 Though these results were non-significant, they add to evidence that suggests antipsychotics may be a clinically useful strategy, particularly for subgroups of patients with severe depression and psychotic symptoms. More recently, a high-quality randomized controlled trial (RCT) found the addition of aripiprazole to venlafaxine resulted in a 44% remission rate compared to 29% in the placebo group in 181 patients with treatment-resistant depression and a mean age of 66 years (odds ratio 2.0; 95% CI 1.1–3.7).37 In older adults, especially in those with a history of cardiovascular disease, antipsychotics should be used cautiously and with careful monitoring to minimize side effects. If there are concerns about cognitive impairment, antipsychotics should be avoided because of the increased risk of stroke.38
There is good evidence that maintenance therapy in patients over 60 years who have achieved remission is effective in preventing relapses and recurrences (number needed to treat = 3.6; 95% CI 2.8–4.8), with comparable efficacy and tolerability seen in SSRIs and tricyclics.39 Given the high risk of relapse in older patients, pharmacotherapy should be continued for at least two years in those who have had two episodes and perhaps indefinitely in those who have suffered three or more.30 There is limited evidence regarding the addition of psychological therapies in older adults to prevent relapse. A 2016 Cochrane review concluded that there was insufficient evidence available regarding psychological therapies preventing a relapse of depression in older people to be able to draw any conclusions or make recommendations.40
There is increasing evidence that collaborative care is an effective non-pharmacological approach to the management of depression. In this model, structured, psychotherapeutic approaches help to increase the effectiveness of pharmacological therapy. In one study, older people with moderate to severe depression who were allocated a nurse who assessed symptoms regularly, co-ordinated care, managed medication, educated and set goals experienced a greater reduction in depressive symptoms than those receiving enhanced usual care.41 A further study showed that the use of collaborative care delivered by a case manager, alongside usual treatment by the GP, was effective in reducing depressive symptoms at 4 months compared to usual treatment by the GP alone, albeit this difference was not maintained in the longer term.42 This mirrors earlier studies which showed that a brief intervention combining psycho-education, support and pharmacotherapy led to improved outcomes for older adults compared to pharmacotherapy alone.43
The evidence for use of formal psychological approaches to treat late-life depression is increasing. Modifications to psychotherapeutic techniques include consideration of the context, maturity of patients and additional factors (e.g. physical illness and cognitive impairment) which may help to improve efficacy when used with older adults.44 A 2015 meta-analysis of psychotherapy for older people with depression showed some evidence, but this varied a great deal depending on the type of control group used.45 Interventions based on cognitive behavioural therapy (CBT) showed that group psychotherapy for older adults with depression had modest efficacy when compared with waiting list controls, which were maintained at follow-up.46 In 2018, a systematic review showed evidence for the use of group therapy, specifically cognitive behavioural and reminiscence-based therapies, for the treatment of depression in older adults.47 A meta-analysis of studies looking at problem-solving therapy (PST) as a treatment for depression in older people found that there was a significant reduction in scores on the Hamilton Rating Scale for depression with the use of PST; however, data on long-term outcomes was lacking.48
There is a limited evidence base for the use of more novel types of therapy in the treatment of depression in older people. An early meta-analysis of 17 studies found that a range of psychosocial treatments for depression in older adults were reliably more effective than no-treatment on self-rated and clinician-rated measures of depression. Therefore, though the evidence is weak, these may still represent important treatment modalities.49 More recent evidence supports the use of Problem Adaptation Therapies for treatment of co-existing depression and cognitive impairment.50 A 2018 meta-analysis showed evidence that the use of exercise as a treatment for depression in older adults is beneficial.51 There is modest evidence for the use of creative arts therapies when delivered by qualified therapists, particularly art, drama and music therapy.52
Electroconvulsive therapy (ECT) is the most effective and rapidly-acting treatment for older people with depression, with remission rates quoted from 60% to 80%; indeed increasing age may be positively correlated with ECT response.53 It is an option both for patients who have not responded to pharmacological therapy and for patients with depressive disorder who need rapid treatment owing to the risk of self-neglect or suicide.54 A recent study comparing ECT to medication (venlafaxine or nortriptyline) in late-life depression found remission rates of 63.8% (30/47) after 6 weeks in the ECT group and 33.3% (27/81) after 12 weeks in the medication group.55
The main risks of ECT are related to the use of a general anaesthetic and, therefore, careful assessment of medical and anaesthetic risks is required, with clinical examination and relevant physical investigations (e.g. ECG, venepuncture and chest X-ray) prior to first ECT, repeated at regular intervals as necessary. Another important consideration is the risk of cognitive side effects, so cognition needs to be carefully monitored prior to and during ECT.56
Bilateral electrode placement is faster and more effective than unilateral. However, it is associated with greater cognitive side effects.57 Therefore, in older patients who are at risk of cognitive impairment, unilateral placement is preferred initially, unless there are urgent clinical reasons which necessitate the use of bilateral ECT. The efficacy of ECT is substantially increased by the addition of an antidepressant medication.58
ECT can be administered on an outpatient basis, provided that the patients are effectively supervised, particularly in the 24 hours following treatment. This is a useful treatment option for patients with severe depression who prefer to avoid hospital admission. Continuation ECT (where the interval between treatments is gradually increased) can reduce and shorten admission for older people with severe depressive disorder and this has been explored further in the PRIDE (Prolonging Remission in the Depressed Elderly) study. In this study, 128 patients who had successfully remitted following treatment three times a week with ultra-brief right-sided unilateral ECT were randomized to either medication-only or flexibly administered maintenance ECT arms.59 At 6 months, only 13% of those in the flexible ECT group had relapsed compared to 20% in the medication arm, and the ECT-treated patients were 5 times more likely to be scored as 1 (‘not ill at all’) on the Clinical Global Impression (CGI) scale.