Author : Kar, Sujita Kumar
Artificial intelligence-based models for augmenting media reporting of suicide: challenges and opportunities
The sensationalised and harmful content of media reporting of suicide is a modifiable risk factor for suicide and suicidal behaviour. Although the World Health Organization (WHO) has published guidelines for responsible media reporting of suicide to prevent suicide contagion, the uptake of these recommendations across media outlets remains limited due to several barriers such as the motivation of stakeholders, inadequate training of media personnel, and a lack of real-time monitoring by the government. In this report, we suggest that artificial intelligence (AI) based models, can be used to address barriers to guideline adherence and improve the quality of media reporting. It is our understanding that the development and implementation of AI-based models or tools can assist in improving adherence to suicide reporting guidelines. We propose a hybrid model that incorporates steps that can be taken at different levels of the media news communication cycle. The algorithmic approach can help in simultaneously processing large amounts of data while also facilitating the design of article structures and placement of key information recommended by media reporting guidelines. The potential benefits of the AI-based model to the various stakeholders and the challenges in implementation are discussed. Given the positioning of responsible media reporting of suicide as a key population-level suicide prevention strategy, efforts should be made to develop and evaluate AI-based models for improving the quality of media reporting in different national or international settings.
Association of executive function, craving and precipitants of relapse in alcohol use disorder: A cross-sectional study
Objective: Alcohol use disorder (AUD) is a global health concern. Patients with AUDs often relapse. Various psychosocial factors, as well as cognitive factors, determine relapse. Failure of response inhibition is often associated with relapse. This study aimed to evaluate the association of craving and relapse precipitants with executive function in AUD.
Materials and methods: The study was conducted in the outpatient setting of a tertiary care hospital in North India (between September 2017 to August 2018) on patients with AUD, who presented with a recent relapse.
A total of 46 adult patients with AUD, who relapsed after a quit attempt were enroled in the study. Cross-sectional assessment of relapse precipitants (by using relapse precipitant inventory), craving (by using the obsessive-compulsive drinking scale (OCDS)), and executive function (EF) (by using the Wisconsin Card Sorting Test (WCST)) was done along with various socio-demographic and clinical variables.
Results: The mean age of onset of alcohol use was 21.48±4.25years and the mean duration of alcohol use was 15.13±7.70 years. The average number of relapses in the study population was 3.59±2.06. There is a significant positive correlation between a negative mood state (as a relapse precipitant) and total relapse score with craving. There is a significant association of relapse and craving with deficits of EF (perseverative and non-perseverative errors). Similarly, lessened cognitive vigilance also significantly correlate with EF deficits resulting in a relapse of AUD.
Conclusion: There is a close association of craving, and relapse with deficits of EF, in AUD. Craving and relapse in AUD may be the result of deficits in EF. Future research addressing the cognitive deficits may help in the prevention of craving and relapse.