Early intervention: Lacks benefit because of lax definition?

At a first pass, the title of this brief piece may seem a tad trenchant. But alas, it is sadly necessary because the major problem with the widely adopted term early intervention (EI), as in EI for psychosis, is that it has become a mantra that is repeated and promoted not on the basis of evidence but largely because it is a truism. The addition of ‘early’ to create the descriptor ‘early intervention’ is inherently attractive because it augments the positive qualities of being proactive by intervening with the culturally valued act of doing something sooner than expected. It also conjures up ideas of delivering treatments that are ‘new’, insinuating that these are appropriate because they are being delivered promptly. In other words, it deftly emphasises the desirable feature of being efficient while saying nothing about the suitability of the intervention being delivered. It is thus understandable that in the context of any debate concerning the merits of EI, the claim that is often made by proponents is ‘who can argue against it?’ The implication being that it is so obviously the correct thing to do that surely no one can oppose it. Well, (yes) we can.

Proponents of EI often fail to recognise that those questioning EI are not challenging the overarching idea per se, but are instead wanting clarification as to how best to apply this approach with greater specificity, so that meaningful effects can be achieved. Key questions include (1) What should EI look like for psychosis? (2) When is the optimal time for various strategies to be implemented? (3) What is the evidence for long-term benefit? (4) What are the risks associated with this approach? Remarkably, after nearly a quarter of a century has passedfollowing the introduction of the EI concept for psychiatry, none of these questions can be adequately answered, which again leads to another important question: why has progress been so slow?

Part of the reason is that these questions are inherently difficult, and adequate answers are only possible with significant advances in understanding. In addition, the nature of psychosis and the causes of schizophrenia remain a mystery. However, another important reason is that many of these questions have simply been set aside. Put another way, they have not been pursued sufficiently, and instead the field has persevered clinically with the broad overarching approach of intervening early. The paradigm has simply not been questioned and has simply been assumed to be appropriate and effective, and thus adopted without much qualification (Moritz et al., 2019). This expedited uptake of EI has been made possible by simply transposing existing treatments that had been used mainly to tackle chronic and entrenched disease to the treatment of purported ‘early stages’ of illness – with the supposition that because they quell acute symptoms they perhaps also have preventive and prophylactic actions (Stain et al., 2019). An additional factor is that purely from a marketing perspective, the dissemination of the EI model has been tremendously successful, that is, EI has been ‘sold’ well (Malhi et al., 2020a). But of all the disciplines within medicine, psychiatry especially should learn from its past – particularly in light of ramifications from the rapid and uncritical acceptance of psychosurgery and insulin coma therapy – and should approach the introduction of any new therapeutic paradigm with due caution.

Transposing treatments

Schizophrenia is a severe, heterogeneous and debilitating illness. It unquestionably has devastating effects, not only on the functioning of the brain and the individual’s social connections within their community, but also on the person’s dignity, their self-worth and sense of self. The illness emerges insidiously, and over time it gradually gains momentum until it often dominates and defines the individual’s existence. The interplay between genetic vulnerability and environmental/lifestyle factors, believed to contribute to the development of schizophrenia, means that parallels can be drawn to other chronic illnesses.

Let us consider the example of cardiovascular disease, where the laying down of atheroma and creeping development of hypertension, that is partly the function of genetic vulnerability, are added to by many lifestyle factors such as smoking and the experience of stress. This can gradually lead to changes in blood vessels that impact various organs and systems within the body and produce a broad range of symptoms. The treatments administered for cardiovascular disease can potentially slow down the progress of these changes and perhaps even reverse some of them. But importantly, these remedial strategies are completely separate from those used to manage additional acute events such as a myocardial infarct. In terms of causation, the latter is also usually the culmination of years of hypertension and the development of atheroma, and the deleterious effects of lifestyle habits and day-to-day stressors, but its management is necessarily very different from the long-term administration of preventive medications such as statins. Why then in the EI of psychosis is there a readiness to adopt medications that are mainly used to treat the severe end of the spectrum and apply them to the management of subtle nascent symptoms with the untested aim of achieving prevention and prophylaxis? What is the rationale? Surely this is an unnecessarily risky strategy that requires new evidence from separate purpose-built studies to support its implementation.

Tentative timing

As with any intervention used to mitigate risk, establishing the appropriate timing is essential. Consider this example: Australia is prone to fires, and in recent years it has experienced devastating losses because of uncontainable bushfires. The best and most obvious strategy is that of prevention, but there is only a relatively small window during which targeted firefighting can hope to contain and prevent the development of an overwhelming bushfire. The strategies used to tackle fire (e.g., depriving it of oxygen, dousing vegetation and trees with water, and backburning bushland to prevent feeding the fire with fuel) can all be used on a small or large scale where appropriate. However, whereas a large raging bushfire cannot be ignored, fire when it first starts can often be difficult to detect, and so for prevention and containment to be effective, knowing where the fire starts is critical, and then getting to it quickly so as to extinguish it before it progresses is vital. In other words, a combination of appropriate timing and reliable identification of the starting site of the fire are critical in bushfire prevention efforts.

The EI paradigm in psychosis appropriately emphasises the need to act and do so promptly. But in order to do so effectively, it is important to identify where the problem lies and at what point and for how long intervention would best be applied. The difference with psychosis is that its origins within the brain, that is, its point of inception, are unknown, both in terms of where and how it begins, and unless the nature of these early changes is known, bathing the whole brain with the neurochemical equivalent of an anti-flagrant may cause more harm than good (Fusar-Poli et al., 2020). This is especially the case if this approach is only likely to be successful if maintained long term.

Proponents of EI often become defensive when questioned about the principles of their advocated approach – assuming that the whole concept is being criticised (Malhi et al., 2020b). After all, who can argue against intervening early? Surely there can be no downside? But this is not the argument being put forward. Expanding a little, the criticisms are the following: What does the EI paradigm inadvertently open the door to? Furthermore, are the treatments targeted? Are they effective? And rather than simply advocating that ‘earlier is better’, is there an ‘optimal window’ that is perhaps governed by age or time? For example, a tipping point in many cancers is that between localised growth and its spread as it metastasises. This is because once a cancer undergoes metastasis, the strategy for its management often changes dramatically. This raises the question as to whether there are similar tipping points within the emergence of psychosis. Unfortunately, sweeping and unquestioning enthusiasm for EI, coupled with its broad uncritical uptake and success in garnering funding, has meant that universal strategies and thinking have been applied, but no advances have been made in terms of answering these more specific questions that would enhance earlier treatment approaches.

Thus, when criticising the definition of EI as ‘lax’, we argue that both more specific terminology and more research are needed to determine not only whether intervening early is a good idea, but also when is it a useful strategy and what are its precise benefits. Fortunately, some efforts are being made in this regard, and some early understanding is gradually emerging. While we too are hopeful in the claims of EI and are motivated by the prospect of improving outcomes for those with debilitating psychiatric illnesses such as schizophrenia, we believe it is important to have voices in the field that continue to question and critique this approach, with the aim of achieving further refinement and development. The undifferentiated promotion and adoption of the EI approach threatens to detract from more nuanced research approaches, and our concern is that just as Icarus’ ill-fated flight was based on a good idea, so too might EI end up being an idea that has a kernel of merit, but because of premature implementation, is unable to be sustained as an effective therapeutic paradigm.