Challenges in detecting and treating depression and a glimmer of hope in suicide prevention

After a summer of bushfires, followed by floods and the continuing COVID-19 pandemic, it is heartening to be able to read some positive findings about a possible suicide prevention programme. School students who participated in a teen mental health first aid course (tMHFA) were better able to recognise suicidality and use appropriate first aid interventions towards peers at risk of suicide. Hart et al. (this issue) conducted a cluster randomised crossover trial in high schools in which high school students either received tMHFA (three 75-minute sessions) or a physical first aid course of similar duration. An assessment of mental health literacy was made prior to the programme, immediately after the programme, and 12 months after training. The participants were assessed on the basis of their responses to a vignette about a fellow student with marked depressive symptoms and ‘ … believes his family would be better off without him’. The students who received tMHFA were more likely to recognise suicidality and apply appropriate first aid interventions (e.g. they asked about suicide, told him they noticed something was wrong with him and wanted to make sure he was okay, and suggested that he tell an adult) than the students who had received the physical first aid course. Remarkably, this suggests that providing even a brief four-session course in mental health first aid may go somewhere towards helping students recognising and providing appropriate first aid to their colleagues who may be expressing suicidal ideation. Such first aid programmes could easily be adopted in high schools equipping students with valuable skills that could lower the risk of suicide in this vulnerable group.

However, it is important to note that while this shows some promise as a universal intervention, particularly encouraging help-seeking, it is of concern that the level of healthcare utilisation among suicide attempters is not optimal, as shown in a study from Austria (Niederkrotenthaler et al., this issue). Niederkrotenthaler and colleagues, using administrative data from lower Austria, examined the patterns of healthcare utilisation of patients diagnosed with the common mental disorders who had attempted suicide and compared them to those who did not attempt suicide. The former, those who had attempted suicide, had lower rates of healthcare utilisation (contacts with general practitioners [GPs]) prior to a suicide attempt. Furthermore, women attempters had high higher levels of contact with psychiatrists following an attempt, whereas male attempters had no increase in contacts with psychiatrists and their level of contact with GPs, in fact, dropped following the attempt. Those who attempted suicide had been prescribed higher levels of psychotropic medications (antidepressants and antipsychotics) and benzodiazepines, especially the male attempters. The authors discuss the reasons for this poor follow-up of such patients, including the point that such patients may be disinclined to seek a consultation or keep follow-up appointments. This is a critical issue for all of us and we need to be rigorous in ensuring such patients attend follow-up visits. The high levels of medication use, accompanied by lack of contact, suggest a lack of therapeutic engagement – perhaps because it is easier to write a prescription than invest time in developing a good therapeutic relationship that requires frequent contact. They also noted that there was a very low use of psychotherapy services following a suicide attempt suggesting once again that this too is an area of marked unmet need.

Antidepressant use across Australia, and worldwide, has been increasing, but, as has been noted previously in this Journal, there has been no discernible decrease in the population prevalence of depression, begging the question: Are antidepressants being prescribed appropriately? Davidson et al. (this issue) examine this in a longitudinal cohort study of antidepressant use in primary care. The Diamond Study is a longitudinal cohort study of patients attending primary care in which 789 adult primary care patients with depressive symptoms were followed up for 2 years every 3 months examining the appropriateness of antidepressant use. Under-treatment (antidepressants not being prescribed when indicated) and overuse of antidepressants (antidepressants prescribed when not indicated) were both found to be prevalent. They found that 41% of the patients were taking an antidepressant for emotional well-being. Among this group, antidepressants were likely to be indicated for use (according to the NICE guidelines) in 59%, suggesting appropriate treatment, but they found that in close to half of the of the assessments (41%), patients were identified as being undertreated. They were not able to identify any specific contributions to either under- or overtreatment other than for patients with one or more chronic illness where antidepressants were prescribed but not indicated.

Depression is frequently comorbid with medical illnesses as highlighted in a timely review by Rosenblat et al. (this issue). They review key issues regarding the bidirectional relationship between medical illness and depression. Correctly identifying de-pression in the medically ill is difficult, in part, because of the overlap in symptoms – but the tools that this paper reviews could assist in making an accurate diagnosis. The authors propose a model for assessing the spectrum of depressive symptoms, ranging from ‘normative’ sadness and adjustment disorders through to major depression, taking note of how the profile of symptoms changes as major depression becomes an important comorbidity. Both pharmacological and psychological treatments of depression in the medically ill are reviewed. With regard to treatments, a collaborative care model is emphasised with a central care manager for the patient to coordinate the management.

One of the problems with depression is that there is no specific diagnostic test for it and instead we rely on our judgement to make the diagnosis that is based on the patient’s history. There have been attempts over the years to identify a biomarker or a laboratory test for depression, dating back to Carroll’s (Carroll et al., 1981) work on the Dexamethasone Suppression Test (DST) for depression and, in this issue of the Journal, two studies are reported. Vascular Endothelial Growth Factor A (VEGF-A) is a growth factor with pro-angiogenic and neurotrophic properties and anti-VEGF-A drugs (used in the treatment of some cancers) have been shown to induce depressive symptoms. This has led to speculation about the levels of VEGF-A in depression with conflicting results found (perhaps a function of different assays used). Rigal et al. (this issue) examined plasma levels of VEGF-A in a consecutive sample of patients with major depression before and after antidepressant treatment. VEGF-A levels were lower in the patients, in comparison with a healthy control group, and the levels of VEGF-A did not significantly change over the course of treatment, or show differences between remitters and non-remitters. Further work on VEGF-A will need to be done to see whether there is any potential for this to be used as a biomarker for depression.

Machine learning was described in an earlier issue of the ANZJP (Cearns et al., 2020) in the context of imaging studies, and in this issue of the journal it has been used to identify a set of biomarkers that could distinguish between patients with either unipolar depression, bipolar depression and healthy controls (Wollenhaupt-Aguiar et al., this issue). The study selected a set of biomarkers that have salience for depression, namely, inflammatory markers, markers of oxidative stress and BDNF, and from this, the authors were able to discern patterns that distinguished unipolar and bipolar depression, bipolar depression from healthy controls and unipolar depression from healthy controls. The sensitivity and specificity of these classifications were reasonable, in the order of 60–70%; however, such findings will need to be replicated in larger samples to determine whether these markers can be used to aid diagnosis in clinical practice.

Anxiety, and the anxiety disorders, are common in pregnant women and in the postpartum period. The treatment of these disorders over the perinatal period is no different from the standard treatments of the anxiety disorders at other times in the life cycle (Andrews et al., 2018), but we know that many women are reluctant to take medication because of concerns about the impact of drugs on the developing foetus or nursing infant. With this in mind, Green et al. (this issue) have developed a six-session manualised Cognitive Behaviour Group Therapy (CBGT) specifically for perinatal anxiety. This brief group programme covers the key elements required for CBT for anxiety disorders and relevant transdiagnostic issues (e.g. depression). A randomised controlled trial in which participants were randomised into CBGT or waitlist control demonstrated the efficacy of this. Women in the CBGT has significant reductions in both self-reported, and clinician rated, anxiety symptoms compared with the control group, and they also had a significant reduction in clinician rated depression, with gains maintained at a 3-month follow-up.

A good clinical handover is necessary for patient safety; unfortunately, this can be less than optimal, especially in those late night phone calls from registrars to discuss patients coming in through the Emergency Department. There is some good news; McKay (this issue) outlines a clear and logical framework for the handover (or baton change) focusing on the ISBAR (Introduction, Situation, Background, Assessment and Recommendation) framework. If registrars were encouraged to use this framework, we would be able to get to the core of the patient’s presen-tation quickly and have a straightforward discussion about management. I encourage you to pass on this excellent paper to your junior staff; perhaps this should be a core part of stage 1 training and an activity that should be entrusted.