Health services research in mental health: The next frontier?

The coronavirus pandemic that we are all dealing with has had the dual effects of increasing community distress and demand for services and simultaneously focusing on models of service delivery. The welcome promise of injection of resources into the mental health sector has served to reinforce the debate around the structure of services. It is therefore timely that this issue of the Journal has its principle focus on health service delivery. Allison et al. (this issue) debate the Trieste model of de-hospitalisation. This was initially widely celebrated as a model of community care. However, the authors note its success is predicated on a low youth population ratio, low rates of drug use and ample social housing and social inclusion. Adequate resourcing of community services is also essential. Allison questions whether our local circumstances parallel those in Trieste and whether we should jump off the Trieste bandwagon and create our own model tailored to our cultural and social contexts.

Given the burden of psychiatric disorders, prevention has become an alluring yet complex goal (Hoare et al., 2020). Looi et al. (this issue) enter this debate arguing that notwithstanding current needs, primary universal selected and indicated prevention of psychosis remain beyond the grasp of our current scientific capacity. Drawing on Cochrane meta-analytic data, they argue that our capacity to predict transition to psychosis is limited and that we currently do not have specific interventions capable of addressing this need (Bosnjak Kuharic et al., 2019). This is not a view shared by McGorry and Nelson (2020) who note over 20 randomised control trials for high-risk participants. They also argue that the control condition in many of these studies is an active psychosocial intervention that may aid the clinical needs of the participants but which might reduce perceived efficacy. Finally, they motivate for a population health perspective that engages individuals at risk of a diversity of phenotypic endpoints. Echoing this perspective, Fusar-Poli et al. (2020) notes advances in the detection and prognosis of people at high risk of psychosis and argues for the need for specialised services and needs-based interventions.

The coronavirus epidemic has also focused service delivery models on telehealth and mobile health. The viewpoint of Byrne et al. (this issue) explores the opportunities and challenges of mobile health technologies in managing mental illness. Biometrics generated by mobile health devices include sleep activity and arousal. These allow continuous monitoring as proxies of psychopathology that have the theoretical capacity of integration into routine care. Challenges include the sensitivity and specificity of such biometrics, their capacity to be perturbed by extraneous factors, issues pertaining to privacy and confidentiality as well as the ever present challenges in integrating rapidly evolving novel technologies into relatively static and inflexible models of current clinical care with high inertia and workload.

Service planning and delivery requires quality estimates of prevalence, disability and burden. This is a particular need for children and adolescents with intellectual disability, highlighted by Buckley et al. (this issue). They note a high 38% prevalence of mental health difficulties, with the most common being attention deficit/hyperactivity disorder, conduct disorder and anxiety disorder.

It is similarly of considerable clinical value to be able to predict a potential clinical trajectory. In a large cohort of people with bipolar disorder from the French FondaMental Advanced Centres of Expertise in Bipolar Disorders, three distinct trajectories were shown over a 3-year follow-up (Godin et al., this issue). The most common trajectory (72%) was a stable pattern of mild functional impairment. A stable pattern of severe impairment was seen in 20%, while 8% had moderate impairment that improved over follow-up. Predictors of a poor outcome included previous hospitalisation, unemployment, childhood maltreatment, residual depression, sleep disturbance, obesity and the need for polypharmacy – some of which are potential intervention targets.

The well-being of medical students and doctors and suicide in this population has been a consistent concern. Vollmer-Conna et al. (this issue) surveyed 151 fourth-year medical students, finding a significant subgroup with impaired mental, psychosocial and physical well-being. A third had at least one current or past psychiatric diagnosis based on the MINI. Maladaptive coping behaviours in a subgroup were evident. The study showed an association between hazardous alcohol consumption, present in 17%, and systemic inflammation, indexed by elevated C-reactive protein. They also replicated the association of impaired heart rate variability and the presence of psychopathology. The conflation of stigma, which impairs help seeking and high stress, complicates intervention in this group. The authors conclude that culture change in the profession is needed aligned with organisational supports for a potentially vulnerable group.

Knowledge of barriers to care is essential to improve access to quality care. Schnyder et al. (this issue) studied barriers to mental health care among children and adolescents (4–17 years). The dominant parent-reported barriers were perceived need for self-reliance, uncertainty about how to access help and cost. Among adolescents, self-reliance and concern regarding what others might think were particular barriers. There was a decrease in perceived unmet need between the two survey points in 1998 and 2013–2014; however, the disparity in perception of receiving sufficient care increased.

There are few disorders that pose a greater challenge to clinical services than that of borderline personality disorder. Walton et al. (this issue) report the results of a randomised controlled effectiveness trial with 161 participants comparing dialectical behaviour therapy and a conversational model. While dialectical behaviour therapy is an established therapy, the conversational model is a psychodynamically informed treatment that was developed specifically for borderline personality disorder. In this two-arm parallel group trial, there were no differences between the treatment arms on measures of suicidal and non-suicidal self-injury, but contrary to expectations, the dialectical behaviour therapy group showed superiority with regard to reduction of depression. This study reinforces the utility of dialectical behavioural therapy in an Australian public health service, especially for those presenting with mood symptoms.

Concluding the health services theme of this issue, Peisah and Goh (this issue) note the high prevalence of psychopathology in intensive care settings and argue for liaison psychiatrists to be attached to intensive care units. In summary, this issue of the Journal highlights the diversity of health service research striving to improve outcomes for mental health disorders.