Detecting suicide risk using knowledge-aware natural language processing and counseling service data.

RATIONALE: Detecting users at risk of suicide in text-based counseling services is essential to ensure that at-risk individuals are flagged and prioritized. OBJECTIVE: The objective of this study is to develop a domain knowledge-aware risk assessment (KARA) model to improve our ability of suicide detection in online counseling systems. METHODS: We obtained the largest known de-identified dataset from an emotional support system established in Hong Kong, comprising 5682 Cantonese conversations between help-seekers and counselors. Of those, 682 conversations disclosed crisis intentions of suicide. We constructed a suicide-knowledge graph, representing suicide-related domain knowledge as a computer-processible graph. Such knowledge graph was embedded into a deep learning model to improve its ability to identify help-seekers in crisis. As the baseline, a standard NLP model was applied to the same task. 80% of the study samples were randomly sampled to train model parameters. The remaining 20% were used for model validation. Evaluation metrics including precision, recall, and c-statistic were reported. RESULTS: Both KARA and the baseline achieved high precision (0.984 and 0.951, shown in Table 2) and high recall (0.942 and 0.947) towards non-crisis cases. For crisis cases, however, KARA model achieved a much higher recall than the baseline (0.870 vs 0.791). The c-statistics of KARA and the baseline were 0.815 and 0.760, respectively. CONCLUSION: KARA significantly outperformed standard NLP models, demonstrating good translational value and clinical relevance.

A Digital Game and School-Based Intervention for Students in Hong Kong: Quasi-Experimental Design.

BACKGROUND: In Hong Kong, with an increasing number of children experiencing mental health issues, there is a need to not only develop innovative interventions but also develop comprehensive prevention interventions so as to reduce their anxiety symptoms and enhance their emotional management and interpersonal relationships. OBJECTIVE: The aim of this study was to determine the effectiveness of The Adventures of DoReMiFa, an integration model of the cognitive-behavioral approach and positive psychology by using digital game-based and school-based mental health enhancement intervention to magnify the social and emotional health and well-being of the school children in Hong Kong aged 9 to 11 years. METHODS: A quasi-experimental design method was used to evaluate this digital game and school-based intervention. The Adventures of DoReMiFa was piloted in 4 primary schools where students were allocated to either an intervention or a control group. The participants were assessed at pre- and postintervention with a 6-month follow-up measuring their mental health knowledge, levels of anxiety symptoms, positive and negative thinking, perspective-taking, and self-esteem. RESULTS: A total of 459 primary school students from 4 primary schools participated in the study. The response rate on the questionnaires answered on the Web was up to 85.1% (391/459). Compared with the control group, the intervention group was found to have significant association with improved mental health knowledge at the time immediately after the intervention (beta=.46; P=.01) and in the 6-month postintervention period (beta=.66; P<.001); for perspective-taking, the intervention group had exhibited a significant improvement 6 months after the completion of the universal program (beta=1.50; P=.03). The intervention, however, was found not to be effective in reducing the rates of anxiety symptoms and negative thinking among the participating students. CONCLUSIONS: The Adventures of DoReMiFa, an integration of a digital game-based and school-based mental health enhancement intervention, was shown to be effective in elevating the knowledge of mental health and promoting perspective-taking in the primary school students of Hong Kong. Although there was insufficient evidence to support a reduction in symptoms of anxiety and negative automatic thoughts, the overall results were still encouraging in that a preventive effect was found, indicating that the program has the potential to enhance the mental well-being of schoolchildren. It also suggests that knowledge enhancement may not necessarily lead to behavior change, and more focused effort may be needed to achieve the translation. The implications and limitations of this study and suggestions for future research were also discussed.

Factors in association with acceptability of A/H1N1 vaccination during the influenza A/H1N1 pandemic phase in the Hong Kong general population.

A random population-based telephone survey (n=301) was conducted among the Hong Kong general population in July 2009. Past history of seasonal influenza vaccination (OR=2.59-3.13) was associated with intention to take up A/H1N1 vaccination under three hypothetical scenarios (provided at HK$200). Adjusting background variables, other significant factors were identified by stepwise models: perceived side effects (OR=0.33), family members' recommendations and friends' acceptability toward the vaccine (OR=2.80-4.74). In contrast to other studies on seasonal influenza and A/H1N1 vaccination, perceived susceptibility and perceived severity related to influenza A/H1N1 were non-significant. Cultural differences may therefore exist.

Acceptability of A/H1N1 vaccination during pandemic phase of influenza A/H1N1 in Hong Kong: population based cross sectional survey.

OBJECTIVE: To investigate the intention of the Hong Kong general population to take up vaccination against influenza A/H1N1. Setting Cross sectional population based anonymous survey. Participants Random sample of 301 adults interviewed by telephone (response rate 80%). MAIN OUTCOME MEASURE: Intention to take up vaccination against influenza A/H1N1 under five hypothetical scenarios: vaccination is free; vaccination per dosage costs less than $HK100 ( pound8; euro9; $13), $HK101-200, or more than $HK200; and no data are available on the efficacy and safety of the vaccine. Results 45% (n=135) of the participants reported that they would be highly likely take up vaccination if it was free. When vaccination incurred a cost, however, the prevalence of uptake decreased: 36% (n=108) would take up vaccination if it cost less than $HK100, 24% (n=72) if it cost $HK101-200, and 15% (n=45) if it cost more than $HK200; and in absence of proved efficacy and safety decreased to 5% (n=14). Moreover, 32% (n=95) considered universal A/H1N1 vaccination unnecessary. Overall, 39% (n=117) of participants believed that A/H1N1 vaccination would prevent the virus being contracted; 63% (n=189) erroneously believed that efficacy of the vaccine had been confirmed by clinical trials, and 16% (n=49) believed that it is necessary for everyone in Hong Kong to take up vaccination against influenza A/H1N1. Conclusions The uptake of vaccination against influenza A/H1N1 by the general population of Hong Kong is unlikely to be high and would be sensitive to personal cost. Evidence about safety and efficacy is critical in determining the prevalence of uptake of vaccination.