Development of Internet suicide message identification and the Monitoring-Tracking-Rescuing model in Taiwan.

BACKGROUND: Suicide messages can be transmitted infinitely online; the Internet is influential in suicide prevention. Identifying suicide risks online via artificial technological advances may help predict suicide. METHODS: We built a classifier that detects open messages containing suicidal ideation or behavior-related words in social media via text mining methods and developed the Monitoring-Tracking-Rescuing model, which links data monitoring and tracking to high-risk suicide rescues. Natural language processing (NLP) techniques such as Long Short-Term Memory and Bidirectional Encoder Representations from Transformers were applied to online posts of common social media sites in Taiwan. This model uses a two-step high-risk identification procedure: an automatic prediction process using NLP to classify suicide-risk levels, followed by professional validation by a senior psychiatrist and a nursing faculty specialized in suicidology. RESULTS: From a dataset containing 404 high-risk and 2226 no- or low-risk articles, the sensitivity and specificity of our model reached 80 %. LIMITATIONS: The model is limited to data platforms that can be "crawled" and excludes suicide-risk content from graphics, video and audio files. Additionally, machine learning does not provide the best recognition rate from complex online messages. Keywords for high-risk suicide in long articles are difficult to interpret using this model. Finally, the model lacks keywords for suicide-protective factors. CONCLUSIONS: Artificial intelligence techniques may help detect and monitor high-risk suicide posts and inform mental health professionals of these posts. Periodic tracking plus manual validation to determine risk levels are recommended to enhance the reliability and effectiveness of Internet suicide-prevention tasks.

Increased risk of hypertension in patients with anxiety disorders: a population-based study.

OBJECTIVE: We designed this study to examine the prevalence and incidence of hypertension and risk factors in patients with anxiety disorders. METHODS: The National Health Research Institute provided a database of 1,000,000 random subjects for study. We obtained a random sample of 766,427 subjects aged ≥18years in 2005. The differences in the prevalence of hypertension between patients with anxiety disorders and the general population in 2005 were tested by multiple logistic regression adjusted for other covariates, including age, sex, insurance amount, region, urbanicity and depression. The differences in the incidence of hypertension between patients with anxiety disorders and the general population according to age group and sex from 2006 to 2010 were tested by a Cox regression adjusted for the other covariates. RESULTS: The prevalence of hypertension in patients with anxiety disorders was higher than that in the general population (37.9% vs. 12.4%, odds ratio, 2.61; 95% confidence interval, 2.52-2.70) in 2005. The average annual incidence of hypertension in patients with anxiety disorders from 2006 to 2010 was also higher than that in the general population (3.63% vs. 1.95%, risk ratio, 1.29; 95% confidence interval, 1.23-1.36). Compared with the general population, patients with anxiety disorders had a higher incidence of hypertension in all age and sex groups. CONCLUSIONS: Patients with anxiety disorders had a higher prevalence and a higher incidence of hypertension than that in the general population. Age, male sex, diabetes, and hyperlipidemia were risk factors for hypertension in patients with anxiety disorders.

Increased risk of hypothyroidism and hyperthyroidism in patients with major depressive disorder: a population-based study.

OBJECTIVE: To determine the prevalence and incidence of hypothyroidism, hyperthyroidism, and risk factors in patients with major depressive disorder (MDD). METHODS: The National Health Research Institute provided a database of 1,000,000 random subjects for health service studies. We identified subjects aged ≥18 years who had at least 1 service claim during 2005 with a primary diagnosis of MDD or with a primary or secondary diagnosis of hypothyroidism or hyperthyroidism. We also compared the incidence of hypothyroidism and hyperthyroidism among patients with MDD and the general population from 2006 through 2010. RESULTS: The prevalence of hypothyroidism in patients with MDD was higher than that in the general population (1.20% vs. 0.30%; odds ratio, 3.08; 95% confidence interval, 2.35-4.03) in 2005. The prevalence of hyperthyroidism was also higher in patients with MDD than in the general population (2.46% vs. 0.79%; odds ratio, 2.77; 95% confidence interval, 2.29-3.35) in 2005. The annual incidence of hypothyroidism was higher in patients with MDD than that in the general population (0.40% vs. 0.13%; risk ratio, 2.47; 95% confidence interval, 2.00-3.06). The annual incidence of hyperthyroidism was also higher in patients with MDD than that in the general population (0.72% vs. 0.32%; risk ratio, 2.06; 95% confidence interval, 1.75-2.43). CONCLUSIONS: Patients with MDD had a higher prevalence and a higher incidence of hypothyroidism or hyperthyroidism than the general population. Female sex was a risk factor for hypothyroidism and hyperthyroidism in MDD.

Increased risk of hypertension in patients with major depressive disorder: a population-based study.

OBJECTIVE: We conducted this study to detect the prevalence, incidence, and risk factors of hypertension in Taiwanese patients with major depressive disorder (MDD). METHODS: The National Health Research Institute provided a database of 1,000,000 random subjects for study. We obtained a random sample of 766,427 subjects, aged ≥18 years in 2005. Study subjects had at least one service claim during 2005 for either outpatient or inpatient care with a primary diagnosis of MDD or with a primary or secondary diagnosis of hypertension and combined with antihypertensive drug treatment were identified. We also compared the incidence of hypertension among patients with MDD and the general population from 2006 through 2008. RESULTS: The 1-year prevalence of hypertension in patients with MDD was higher than that in the general population (21.21% vs. 13.28%, risk ratio, 1.22; 95% confidence interval, 1.17-1.28). Compared with the general population, patients with MDD had a higher prevalence of hypertension in all age, sex, insurance amount, region, and urbanicity groups. The average annual incidence of hypertension in patients with MDD from 2006 to 2008 was higher than that in the general population (3.96% vs. 2.90%, risk ratio, 1.19; 95% confidence interval, 1.08-1.31). Compared with the general population, patients with MDD had a higher incidence of hypertension in all age, and male groups. CONCLUSIONS: Patients with MDD had a higher prevalence and a higher incidence of hypertension than that in the general population. Age, male gender, and antipsychotic use were risk factors for hypertension among patients with MDD.

Prevalence of diabetes in patients with major depressive disorder: a population-based study.

OBJECTIVE: We conducted this population-based study to detect the prevalence and associated factors of diabetes in patients with major depressive disorder (MDD) in Taiwan. METHODS: The National Health Research Institute provided a database of 1,000,000 random subjects for health service study. We obtained a random sample of 766,427 subjects 18 years or older, in 2005. Study subjects who had at least one service claim during 2005 for either outpatient or inpatient care, with a primary diagnosis of MDD or with a primary or secondary diagnosis of diabetes, were identified. RESULTS: The 1-year prevalence of diabetes in patients with MDD was higher than that in the general population (11.65% vs 6.53%; odds ratio, 1.53; 95% confidence interval, 1.39-1.69) in 2005. Compared with the general population, patients with MDD had a higher prevalence of diabetes in all age groups, except age 18 to 29 years; among men and women; among all insurance amount groups; among those living in the northern and southern regions; and among residents living in urban, suburban, and rural areas. A higher prevalence of diabetes in patients with MDD was associated with increased age, use of antipsychotic agents, use of mood stabilizers, and residence in suburban areas. CONCLUSIONS: Patients with MDD had a much higher prevalence of diabetes in young adult age group and in men than in the general population. Consequently, we must emphasize prevention, early detection, and adequate treatment of diabetes in patients with MDD.