Comparison of unweighted and item response theory-based weighted sum scoring for the Nine-Questions Depression-Rating Scale in the Northern Thai Dialect.

BACKGROUND: The Nine-Questions Depression-Rating Scale (9Q) has been developed as an alternative assessment tool for assessing the severity of depressive symptoms in Thai adults. The traditional unweighted sum scoring approach does not account for differences in the loadings of the items on the actual severity. Therefore, we developed an Item Response Theory (IRT)-based weighted sum scoring approach to provide a scoring method that is more precise than the unweighted sum score. METHODS: Secondary data from a study on the criterion-related validity of the 9Q in the northern Thai dialect was used in this study. All participants were interviewed to obtain demographic data and screened/evaluated for major depressive disorder and the severity of the associated depressive symptoms, followed by diagnosis by a psychiatrist for major depressive disorder. IRT models were used to estimate the discrimination and threshold parameters. Differential item functioning (DIF) of responses to each item between males and females was compared using likelihood-ratio tests. The IRT-based weighed sum scores of the individual items are defined as the linear combination of individual response weighted with the discrimination and threshold parameters divided by the plausible maximum score based on the graded-response model (GRM) for the 9Q score (9Q-GRM) or the nominal-response model (NRM) for categorical combinations of the intensity and frequency of symptoms from the 9Q responses (9QSF-NRM). The performances of the two scoring procedures were compared using relative precision. RESULTS: Of the 1,355 participants, 1,000 and 355 participants were randomly selected for the developmental and validation group for the IRT-based weighted scoring, respectively. the gender-related DIF were presented for items 2 and 5 for the 9Q-GRM, while most items (except for items 3 and 6) for the 9QSF-NRM, which could be used to separately estimate the parameters between genders. The 9Q-GRM model accounting for DIF had a higher precision (16.7%) than the unweighted sum-score approach. DISCUSSION: Our findings suggest that weighted sum scoring with the IRT parameters can improve the scoring when using 9Q to measure the severity of the depressive symptoms in Thai adults. Accounting for DIF between the genders resulted in higher precision for IRT-based weighted scoring.

Services for depression and suicide in Thailand.

Depression, together with suicide is an important contributor to the burden of disease in Thailand. Until recently, depression has been significantly under-recognized in the country. The lack of response to this health challenge has been compounded by a low level of access to standard care, constraints on mental health personnel and inadequate dissemination of knowledge in caring for people with these disorders. In the past decade, significant work has been undertaken to establish a new evidence-based surveillance and care system for depression and suicide in Thailand that operates at all levels of health-care provision nationwide. The main components of the integrated system are: (i) community-level screening for depression in at-risk groups, using a two-question tool; (ii) assessment of the severity of depression using a nine-question scale; (iii) diagnosis and treatment by general practitioners; (iv) psychosocial care provided by psychiatric nurses; (v) continuous care for relapse and suicide prevention; and (vi) promotion of mental well-being and prevention of depression in at-risk populations. Factors such as appropriate financial mechanisms, capacity-building programmes for health-care workers, and robust treatment guidelines have contributed to the success and sustainability of this comprehensive surveillance and care system. By 2016, more than 14 million people at risk had been screened for depression and received mental health education; more than 1.7 million people with depression had received psychosocial interventions; 0.7 million diagnosed patients had received antidepressants; and 0.8 million were being followed up for relapse and suicide prevention. The application of this surveillance and care system has led to an enormous increase in the accessibility of standard care for people with depressive disorders, from 5.1% of those with depressive disorders in 2009 to 48.5% in 2016.

Time to relapse and remission of bipolar disorder: findings from a 1-year prospective study in Thailand.

BACKGROUND AND METHODS: This study aimed to determine time to relapse and remission of mood episodes in Thai patients with bipolar disorder (BD). The Thai Bipolar Disorder Registry was a multicenter, prospective, naturalistic, observational study conducted in Thailand. Participants were adult inpatients or outpatients with Diagnostic and Statistical Manual of Mental Disorders bipolar disorder. The diagnosis of bipolar disorder, current psychiatric comorbidity, mood relapse, and mood remission were determined by using the Mini International Neuropsychiatric Interview. Relapse and remission were assessed every 2 months. RESULTS: Of 424 BD participants, 404 (95.3%) were BD I, and 258 (60.8%) were female. At entry, 260 (61.3%) had recovered, and 49 (11.6%) were recovering. During 1-year follow-up (381.7 person-years), 92 participants (21.7%) had 119 relapses or 0.31 (95% confidence interval 0.25-0.35) episodes per person-year. Among 119 relapses, 58 (48.7%), 39 (32.7%), and 21 (17.6%) of them were depressive, hypomanic, and manic episodes, respectively. Using the Kaplan-Meier method, we found that 25% of the participants relapsed in 361 days. Of the 400 participants who reached remission, 113 (28.2%) had mood relapses. Of 173 mood events accountable for remission analysis, the median time to remission was 67.5 days (72.5 days for depressive episodes versus 58.0 days for manic episodes, log rank P = 0.014). CONCLUSIONS: The 1-year relapse rate in Thai patients with BD was 21.7% or 0.31 episodes per person-year. About one-fifth of recovered patients had mood relapses within 371 days. On average, a mood episode would remit in 67.5 days.

Patterns of alcohol dependence in Thai drinkers: a differential item functioning analysis of gender and age bias.

In Caucasians, the patterns of alcohol use disorders in women and adolescents are likely to be different from those in men and adults, respectively. The authors examined these differences in a Southeast Asian sample of Thai people living in communities. A two-parameter logistic model of the IRT log-likelihood-ratio (IRTLR) test for differential item functioning (DIF) procedure was used. Participants were a subsample of 3718 current drinkers participating in the 2008 Thai National Mental Health Survey (n=17,140). The 1-year prevalence rates of alcohol dependence were 1.4% in women and 13.7% in men. Based on the Mini International Neuropsychiatric Interview (MINI), alcohol dependence and abuse module, all current drinkers were interviewed for a yes/no response to each of seven alcohol dependence criteria. Confirmatory factor analysis suggested a single-factor model of alcohol dependence criteria (χ2=211.51, RMSEA=0.06, SRMR=0.03 and CFI=0.96). Compared with 3174 men, 544 women had a significantly higher threshold estimate for quit/control problems and a lower threshold value for drinking despite physical/mental problems (b parameter difference of 0.25 and -0.30, respectively). Thai adolescents (n=272) and Thai adults (n=3446) had no statistically significant DIF on any criterion. The criterion of time spent drinking had significantly high discrimination estimates in women, men, adolescent and adults (a parameters of 2.50, 2.08, 2.33 and 2.16, respectively). Gender bias on alcohol dependence criteria can be found in Thai drinkers. Time spent drinking may be the most useful criterion for discriminating the severity of alcohol dependence across age and gender groups of Thai drinkers.

Development and validity of two-question screening test for depressive disorders in Northeastern Thai community.

OBJECTIVES: (1) To develop a screening test for depressive disorders which is simple and short. (2) To determine the diagnostic accuracy of this screening test in Northeastern Thai community. METHODS: From August 2006 through October 2006, two phases of tool development were conducted. The first phase, a two-question screening test of depressive disorders was developed by selecting the dialect words to express depressive mood among Northeastern Thai population. The first item represented depressive mood and the second item represented loss of interest or pleasure in doing things. The second phase, a cross-sectional criterion standard validation study was conducted using stratified cluster randomization to identify 1002 subjects ≥18 years of age in 17 villages in nine districts of Yasothorn province. The two-question screening test was used to identify subgroups of subjects positive of both, positive of either one, and negative of both items. Psychiatrists then determined their diagnoses of mood disorders by administering a structured psychiatric examination (M.I.N.I.) that employed Diagnostic and Statistical Manual of Mental Disorder Text Revision (DSM-IV-TR). Data analysis used STATA 8.0 to calculate the sensitivity, specificity, positive predictive value, negative predictive value, percent of correct classified, and prevalence of depressive disorders with 95% confidence interval. RESULTS: Of 1002 subjects, female were more predominant (58%), with mean age of 41.5 years. The overall time point rate of depressive disorders was 5.9% (95% CI=4.3-7.1%). Major depressive disorder 3.7% (95% CI=2.5-4.9%) and dysthymia 2.4% (95% CI=1.5-3.3%) occupied the preceding position. The first question of screening test was more positive 3 times than the second one. It showed a high sensitivity of 96.5% (95% CI=95.4-97.6%) but low specificity of 45.6% (95% CI=42.5-48.7%). The second question showed a low sensitivity of 71.9% (95% CI=69.2-74.7%) but high specificity of 84.1% (95% CI=81.9-86.4%). When either one of two questions was positive, this test showed high sensitivity of 96.5% (95% CI=95.4-97.6%) and the positive likelihood ratio was 1.74 (95% CI=1.61-1.88). If both of two questions were positive, it showed high specificity of 85.1% (95% CI=82.9-87.3%) and the positive likelihood ratio was 4.82 (95% CI=3.86-6.02). CONCLUSION: Two questions for screening of depression would detect most cases of depressive disorders in the Thai community which has the advantage of brevity, high sensitivity and specificity.