Zheng L, Wang O, Hao S, Ye C, Liu M, Xia M, Sabo AL, Markovic L, Stearns F, Kanov
L, Sylvester KL, Widen R, McElhinney DB, Zhang W, Liao J, Ling XB
Development of an early-warning system for high-risk patients for suicide attempt
using deep learning and electronic health records
Transl Psychiatry 2020;10(1):72
The paper of Zheng et al. reports a study focusing on tackling suicide attempt. It
is a public health focused paper combining health informatics and biostatistics methods
and which are applied to stratify patients within a cohort in order to identify those
at risk of suicide. They developed an early-warning system for high-risk suicide attempt
patients through the design and implementation of a population-based risk stratification
surveillance system. A Deep Neural Network (DNN) model was trained for the prediction,
stratification, and calibration. Then a Local Interpretable Model-agnostic Explanations
algorithm was utilised to interpret the risk stratification results. As of result,
a total of 117 features were significant in the predictive model; the DNN on the EHR-based
data, enabled finding that suicide attempts patients were more likely to be gin: age
groups of 6–54, diagnosed mental health conditions or pain, to have suicide attempts,
treated by psychotropic medications, and have open wounds or injuries due to unspecific
reasons. The findings of the study enable early interventions and appropriate treatments
to mitigate suicide risk.
Roope LSJ, Tonkin-Crine S, Herd N, Michie S, Pouwels KB, Castro-Sanchez E, Sallis
A, Hopkins S, Robotham JV, Crook DW, Peto T, Peters M, Butler CC, Walker AS, Wordsworth
S
Reducing expectations for antibiotics in primary care: a randomised experiment to
test the response to fear-based messages about antimicrobial resistance
BMC Med 2020;18(1):110
The aim of this study is to test the likely impact of fear-based messages, with and
without empowering self-management elements, on patient consultations or antibiotic
requests for influenza-like illnesses, using a randomised design. To do so, they relied
on the use of an on-line survey of adult members of the UK general public. They proceed
to the randomisation of patients, to receive one of three different messages about
antibiotics and Antimicrobial Resistance. As of results, they have been able to demonstrate
that empowering patient could help to better involve them in their care process and
make a better decision regarding Antibiotics related issues. The study concludes that
while fear-only messages could be effective in public campaigns to reduce inappropriate
an- tibiotic use, they should be combined with messages empowering patients to self-manage
symptoms effectively without antibiotics.
Degeling C, Carter SM, van Oijen AM McAnulty J, Sintchenko V, Braunack-Mayer A, Yarwood
T, Johnson J, Gilbert GL
Community perspectives on the benefits and risks of technologically enhanced communicable
disease surveillance systems: a report on four community juries
BMC Med Ethics 2020;21(1):31
The paper focuses on assessing how routinely using Whole Genome Sequencing (WGS) and
Big Data technologies to capture more detailed and specific personal information could
be perceived among communities in two demographically different Sydney municipalities
and two regional cities in New South Wales, Australia (western Sydney, Wollongong,
Tamworth, eastern Sydney). Four community juries were created by recruiting participants
at each study site by an independent professional research service. The aim was to
elicite the views of well-informed community members on the acceptability and legitimacy
of making pathogen WGS and linked administrative data available for public health
research using this information in concert with data linkage and machine learning
to enhance communicable disease surveillance systems. Participants across all four
events strongly supported the introduction of data linkage and pathogenomics to public
health research under current research governance structures. This study demonstrates
that when public is well informed, here a jury proxy is used, they are likely to support
routine collection, linkage and use of administrative and pathogenomics data for the
purposes of public health research.
