The public and professionals require accessible summaries of the best research evidence and most may read no more than the press coverage or abstract; if captivated or surprised by the findings, they may proceed to delve into the full text of the research publication. This might help clinicians to improve evidence-based care and researchers to inform their work and plan future work, and for patients and the public the information may provide more personal guidance on care based on evidence. However, researchers and clinicians working in communities, in primary and secondary care apply the findings in particular local contexts and with specific cohorts of complex symptoms and comorbid illnesses; every patient will have their personal worries or concerns informed by their family histories and their illness perceptions about what causes or cures an illness. Even though in 2016 the BJPsych had over 22 600 citations in peer-reviewed journals, this wider usage of research is poorly captured by conventional measures of impact and our metrics.
I reviewed the most ‘popular’ papers of 2016 and quickly encountered a dilemma. What constitutes popularity and value to both the public and scientific community? Should this judgement be based on citations in peer-reviewed scientific journals, despite the concerns expressed about such processes in the scientific community; or should we use more social media or web-based metrics. For example, Altmetric scores capture social media (as well as news media) prominence. Readers and authors can examine more metrics on our website: access to the abstract and the full text (either as full-text HTML file or as PDF) of papers. On the basis of these measures, looking at the three most cited research papers of 2016 in (as on 1 Feb 2017), the top of the list is Brunoni et al's patient-level review and sham controlled meta-analysis of transcranial direct current stimulation for acute depression (counting Altmetric score (in italic), abstract views, full-text views, and PDF views respectively: 34, 8106, 683 and 664).1 This showed depression improvement, remission and dose–response relationships. The second most cited research paper was Esscher et al's study of suicide during pregnancy and for a year after birth (notching up: 3, 3774, 258, 308);2 this paper revealed that antenatal care did not fully capture psychiatric risks, and 26 of 103 women who died had no documented psychiatric care, whereas 20 women had a post-partum care plan that addressed their mental health. This finding has implications for the organisation and planning of interdisciplinary antenatal care, involving both providers of health services but also public health agencies. Winkler et al's review of deinstitutionalisation and its impacts on homelessness or imprisonment ranked third most cited (with 19, 5407, 520, 360).3 The findings challenged prominent views that homelessness and imprisonment were a consequence of closure of large psychiatric hospitals. There ensued a healthy correspondence about possible alternative explanations, and how to respond in the community. Homelessness remains a major priority given the levels of poor mental health and premature mortality in homeless populations.4,5 Although the three papers garnering most citations make relevant and important discovery contributions, contrast these with marginally less cited but important and startling new findings captured in the social media profiles of the following three papers: Crawford et al's survey of patient experiences reported on the adverse effects of psychological therapies (ranked 8 in the top 10 cited papers of 2016, with respective scores of 213, 15971, 1278, 806).6 Similarly remarkable papers include Hirvikoski et al's study of premature mortality in autism (392, 19382, 1451, 1180)7 and Palmier-Claus et al's study of childhood adversity and bipolar disorder (357, 7876, 275, 500).8 The social and news media presence perhaps reflects the absence of good data and the need to address more immediate clinical concerns but also the concerns of patients and families struggling with childhood mental illnesses.
Neither the social media scores nor the online reads relate easily to citations. There are diverse values and types of knowledge sought by our readers, authors, and members of the Royal College of Psychiatrists and other learned societies. Immediate salience to the public, clinicians and scientists may drive the social media presence and levels of online access; however, citations reflect longevity and discovery not for implementation purposes but for building upon the total knowledge base and what is known about the causes of and recovery from mental illnesses. The latter requires a careful read and awareness of the knowledge horizon and whether a new research paper reinforces or calls into question existing practice and evidence of effectiveness. BJPsych now also encourages analysis and debate articles that bring readers up to speed on conflicting or questionable perspectives where there is not enough high-level evidence to definitively recommend one intervention above another, nor to indicate that one is inferior to another, thus permitting more choice for patients. Interactions with readers and ongoing discussion about findings are important indicators of scientific and clinical advances, so you are encouraged to make good use of rapid eLetters to comment on data, refine interpretations or raise questions about published research. The democratisation of knowledge involves a suitably disputatious but respectful exchange of views, ultimately to improve public mental health.
Furthermore, in order to improve access to the latest evidence for the widest audience, we encourage researchers to be more explicit about public and patient engagement, critiques, and impacts in their research reports. The Editorial Board is also testing ways of conveying research findings to the wider public through plain English summaries of the original research, encouraging discussion and sharing of opinions and perspectives through social and digital media. For example, in partnerships with the Mental Health Foundation and The Mental Elf, we are testing out these methods with podcasts on traumatic symptoms9 and on depression,10 and a mental health blogger's analysis of BJPsych papers on weekend admissions.11
This month's BJPsych provides more opportunities to collectively build the evidence and knowledge base for improved care. For schizophrenia, cognitive–behavioural therapy (CBT) and treatment as usual (TAU) combined prove effective compared with TAU alone (Guo et al, pp. 223–229), but this is important as TAU in China is not as comprehensive, with fewer clinicians and services (except perhaps in urban areas) compared with higher-income countries. Thus even in this context CBT seems helpful. A review of patient-level data shows that CBT for depression is effective irrespective of the baseline levels of depression, informing clinicians' and patients' decision-making (Furukawa et al, pp. 190–196). Depression appears to be equally common and in both first- and second-generation immigrants (Mindlis & Boffetta, pp. 182–189), suggesting that migration status should be considered in assessment and recovery plans. Patients receiving psychotropic medication who later present with self-harm appear to be less likely to have co-ingested alcohol, which should improve their outcome given that alcohol is a risk factor for adverse medical outcomes (Chitty et al, pp. 203–208), and mood stabilisers and antipsychotic medication reduce the chances of later admission among people with bipolar disorder (Joas et al, pp. 197–202). Two studies report on neurobiological correlates of auditory hallucinations (Zhuo et al, pp. 209–215; Ramsay et al, pp. 216–222), while Kelleher & DeVylder's findings (pp. 230–231) call into question the common assumption that people carrying a diagnosis of borderline personality disorder are more likely to have auditory hallucinations compared with people with common mental disorders. I welcome your communications and critiques on published research; only with this scrutiny, counter-narrative, and engagement can we hope to achieve our aspirations for better care for mental illnesses and to improve population mental health.
- © The Royal College of Psychiatrists 2017.