Was much more proper to focus on dilemma gamblers (SOGS 3) than the

Was far more suitable to focus on dilemma gamblers (SOGS 3) than the modest group of pathological gamblers. Also, non-gamblers had been separated into their own group. Age of onset of gambling was dichotomised into two SCH 58261 site groups (<18 years, 18 years) based on the age limit for legal gambling in Finland. Health- and well-being related correlates were also investigated. General health was inquired using a question: `How is your general health at present?'. Five response options were dichotomised into two groups: 1) average, good or somewhat good and 2) bad or somewhat bad. Mental health was assessed using the Mental Health Inventory, which comprised five items: nervousness, the blues, jollity, calmness and happiness (MHI-5) [28]. The MHI-5 has a 6-point Likert scale (range 1?). The total MHI-5 scores were calculated by summing up the score of each item, with sums (range 4?0) rescaled to 1?00. A score of 52 or less was used to indicate clinically significant mental health problems, as recommended by Berwick and colleagues [29]. The MHI-5 seems to be an adequate screen for some anxiety disorders (generalised anxiety disorder, panic disorder, obsessive compulsive disorders), but not others, especially phobias [30]. In our study, Cronbach's alpha for the MHI-5 was 0.77. Loneliness was inquired using a question: `Do you feel lonely?'. Five response options for loneliness were dichotomised into two groups: 1) never or very rarely and 2) sometimes, often or all the time. Smoking was inquired using the question: `Have you smoked during the past 12 months?'. Three response options for smoking were dichotomised into two groups: 1) daily smoking and 2) occasionally or not at all. Alcohol consumption was measured using a 3-item version of the Alcohol Use Disorders Identification Test (AUDITC) [31]. The AUDIT-C appears to be practical and valid screen for heavy drinking and/or active alcohol abuse or dependence [31]. Total score for the AUDIT-C was counted by summing the points (range 0?) for each item and using the cut-off points recommended by Sepp?(2010) to define risky drinking among Finnish males (6 points) and females (5 points) [32]. In our study, Cronbach's alpha for the AUDIT-C was 0.61. The Finnish versions of the instruments were translated in collaboration with qualified translators and an expert panel. All instruments were pilot tested (N = 30).Statistical analysesThe data were analysed using SPSS 21.0 software (SPSS, Inc., Chicago, IL, USA). Descriptive statistics includedSalonen et al. BMC Public Health 2014, 14:398 http://www.biomedcentral.com/1471-2458/14/Page 4 offrequencies, percentages, means and standards deviation (SD). First, the gender proportions for CSOs were calculated (Table 1). Then, correlates were examined within genders: male CSOs were compared with male nonCSOs, and female CSOs were compared with female non-CSOs (Tables 2 3). Statistical significance (p) was determined by the chi-squared Test (>2 groups) and Fisher’s Exact Test (two groups), and Binary CJ-023423 cost Logistic Regression Evaluation. Missing values have been not replaced except whilst producing the CSO variable: missing values were integrated into the non-CSO category. Two multivariate models with the correlates are presented (Table three). All variables had been integrated in Model 1 simultaneously. Distinct combinations with the above correlates had been tested while producing Model two. The poorest correlates were dropped. The best fitting model, Model 2, was selected, employing statistical sig.Was extra suitable to concentrate on challenge gamblers (SOGS 3) than the smaller group of pathological gamblers. Furthermore, non-gamblers had been separated into their own group. Age of onset of gambling was dichotomised into two groups (<18 years, 18 years) based on the age limit for legal gambling in Finland. Health- and well-being related correlates were also investigated. General health was inquired using a question: `How is your general health at present?'. Five response options were dichotomised into two groups: 1) average, good or somewhat good and 2) bad or somewhat bad. Mental health was assessed using the Mental Health Inventory, which comprised five items: nervousness, the blues, jollity, calmness and happiness (MHI-5) [28]. The MHI-5 has a 6-point Likert scale (range 1?). The total MHI-5 scores were calculated by summing up the score of each item, with sums (range 4?0) rescaled to 1?00. A score of 52 or less was used to indicate clinically significant mental health problems, as recommended by Berwick and colleagues [29]. The MHI-5 seems to be an adequate screen for some anxiety disorders (generalised anxiety disorder, panic disorder, obsessive compulsive disorders), but not others, especially phobias [30]. In our study, Cronbach's alpha for the MHI-5 was 0.77. Loneliness was inquired using a question: `Do you feel lonely?'. Five response options for loneliness were dichotomised into two groups: 1) never or very rarely and 2) sometimes, often or all the time. Smoking was inquired using the question: `Have you smoked during the past 12 months?'. Three response options for smoking were dichotomised into two groups: 1) daily smoking and 2) occasionally or not at all. Alcohol consumption was measured using a 3-item version of the Alcohol Use Disorders Identification Test (AUDITC) [31]. The AUDIT-C appears to be practical and valid screen for heavy drinking and/or active alcohol abuse or dependence [31]. Total score for the AUDIT-C was counted by summing the points (range 0?) for each item and using the cut-off points recommended by Sepp?(2010) to define risky drinking among Finnish males (6 points) and females (5 points) [32]. In our study, Cronbach's alpha for the AUDIT-C was 0.61. The Finnish versions of the instruments were translated in collaboration with qualified translators and an expert panel. All instruments were pilot tested (N = 30).Statistical analysesThe data were analysed using SPSS 21.0 software (SPSS, Inc., Chicago, IL, USA). Descriptive statistics includedSalonen et al. BMC Public Health 2014, 14:398 http://www.biomedcentral.com/1471-2458/14/Page 4 offrequencies, percentages, means and standards deviation (SD). First, the gender proportions for CSOs were calculated (Table 1). Then, correlates were examined within genders: male CSOs were compared with male nonCSOs, and female CSOs were compared with female non-CSOs (Tables 2 3). Statistical significance (p) was determined by the chi-squared Test (>2 groups) and Fisher’s Precise Test (two groups), and Binary Logistic Regression Evaluation. Missing values were not replaced except though producing the CSO variable: missing values have been included into the non-CSO category. Two multivariate models with the correlates are presented (Table 3). All variables had been incorporated in Model 1 simultaneously. Various combinations of the above correlates were tested when generating Model 2. The poorest correlates were dropped. The most beneficial fitting model, Model 2, was selected, employing statistical sig.