The researcher intended to summarize whether there is any association between socioeconomic factors that lead to suicide rates in Thailand. Models of these factors may have been made in previous studies conducted in developing countries, but the context is definitely distinct from Thailand where the vast majority of people work in the agricultural sector so the employment and income of Thai agrarians are included in the model. In addition, studies that collect the macro-level of data are rare. A previous similar study was conducted in Thailand, but it was outdated since this knowledge was generated, and it did not explore more specifically the factors of cultural, social, and economic differences among regions.
I would like to highlight one of the risk factors that is most associated with suicide rates, namely the percentage of elderly people (age above 60). The linear regression model produced a positive coefficient with a value close to 1 (almost perfect), which was interpreted that if the value of one variable increases, it is likely that the value of the other variables would also increase. In this case, provinces with a higher percentage of elders had higher suicide rates and it ascertains the previous descriptive analysis that explained the trend of increasing suicides among people age over 40 years.
Although the Thai suicide research did not explain causal relationships and did not predict trends of suicide in the future, statistical modeling with linear regression in epidemiological research explains the underlying mechanism of how predictors are associated with suicide problems as the outcome. With the multiple regression model, we can see the variation of the strength of independent variables to the outcome, as well as what is the association’s direction among the two variables. Back to the results about larger rates of suicide in provinces with a higher percentage of elderly people, this linear regression model helps spatial epidemiology to map the vulnerability of each province. The vulnerability map is useful for the government in prioritizing intervention program.