Pinxten WJL et all. Excellent reliability and validity of the Addiction Medicine Training Need Assessment Scale across four countries. Journal of substance abuse treatment. 2019 Apr 1;99:61-6. RAW DATA SET
Description
Background: This data set was used to study the reliability and validity of the Addiction Medicine Training Need Assessment Scale across four countries: The Netherlands, Indonesia Lithuania Ireland as published in the Journal of substance abuse treatment. 2019 Apr 1;99:61-6. For a detailed explanation, we refer to the original article. The aim of this study was to evaluate the psychometric properties of the AM-TNA Scale. Methods: In a cross-sectional study in 4 countries the AM-TNA was distributed in standard training settings. The scores of 403 respondents were used to assess the factor structure using explorative factor analysis. Reliability was tested using Cronbach's Alpha, ANOVA determined the discriminative validity. Results: Validity: factor analysis revealed a two-factor structure: One on providing direct patient treatment and care (Factor 1: clinical) and one factor on facilitating/supporting direct patient treatment and care (Factor 2: non-clinical) AM competencies and a cumulative 76% explained variance. Reliability: Factor 1 α = 0.983 and Factor 2: α = 0.956, while overall reliability was (α = 0.986). The AM-TNA was able to differentiate training needs across groups of AM professionals on all 30 addiction medicine competencies (P = .001). Conclusions: In our study, the AM-TNA scale had a strong two-factor structure and was proven to be a reliable and valid instrument. The next step should be testing external validity, strengthening discriminant validity and assessing the re-test effect and measuring changes over time. Explanation of the RAW DATA Excel sheet. Colum A. ID: #2-51 NL NVV Dutch Addiction Physicians, #52-88: Psychiatry residents (except 66 and 67 psychiatrists) # 89-118: Addiction Physicians in training, #119-145 Indonesian addiction specialists (graduated Psychiatrists), # 146- 221: Indonesian addiction physicians in training, #222-290: Lithuanian Psychiatry residents following addiction specialization, # 291-404 Irish GPs in training. Column B. 1: The Netherlands, 2: Indonesia, 3: Lithuania and 4: Ireland. Column C. 1: Male, 2 Female. Column D and E: 2-290 age in years 1: Younger than 30 years, 2: Between 31 and 50 years 3. Older than 50 years. Column F: #2-51: all Dutch established Addiction professionals, #52-66 all Psychiatry residents, #67 and 69 psychiatrists, #69-88 AIOS Psychiatry Residents in training, #89-118 VIOS: Addiction Physicians in training, #119-145: Established Indonesian addiction professionals, #146-221 Indonesian Addiction specialist in training (99 missing value) 1: addiction specialist, 2: Addiction specialist in training, 3: sometimes patients with SUD 4: never work with patients with SUD, # 222-290 Psychiatrists in training,#29 1-204 Irish GP's in training. Colum G-AJ all 30 competencies scored in a Likert scale as explained in the original article Table 2.
Files
Steps to reproduce
Study design: A cross-sectional, comparative design to assess the psychometric aspects of the AM-TNA. Countries and participants: In order to assess whether the instrument measures professional AM training need differences, we included participants from The Netherlands, Lithuania, Indonesia and Ireland because these countries strongly differ in AM training programs. A convenience sampling approach was used: adding to a total of 428 respondents. The instrument: The AM-TNA was developed in 2010 in Bandung, Indonesia as a tool to develop the Indonesian Addiction Medicine Course (ISCAN) (Pinxten, De Jong, et al., 2011. The AM-TNA is a 30 items questionnaire containing self-reported perceptions of the proficiency on specific AM core competencies, using a five-point Likert scale: from not at all proficient (1) to fully proficient (5). The AM-TNA questionnaire covers a variety of competencies, including the skills in three professional domains: to assess substance use, to start treatment and to maintain treatment of patients with a substance use disorder. The AM-TNA is a paper and pencil questionnaire and completion of the questionnaire takes about 10 min. Analysis: Respondents with missing values in the gender section (NL: 1 and Ire: 7) were excluded for analysis while missing values in the age section were imputed through average value per country (Ire: 1). Because Lithuania only reported age categories, all other age data were subsequently transformed into the same age categories. In order to assess the overall psychometric value of the instrument, we required a sufficient sample size for valid Explorative Factor Analysis (EFA) and merged the 4-country data (raw data set: n = 428, after cleaning and imputation: n = 403). According to Matsunaga (2010), this sample size earns the qualification GOOD. Using descriptive statistics this merged dataset served to analyze demographics, including age and gender. The factor structure of the AM-TNA was analyzed through EFA. With 30 competencies/variables we opted to use the Kaiser's criterion (all factors with eigenvalues > 1 retained) and, because factors are expected to be independent, a varimax rotation to maximize the dispersion of loadings within factors. Reliability was measured using Cronbach's Alpha. Discriminant validity was established through analysis of variance (ANOVA), comparing training needs across all professionals. For analysis of normal distribution, we will apply the Z-test for skewness and kurtosis. SPSS, version 23, was used for the analyses Further details can be found in the attached article The main author can always be contacted: lpinxten@gmail.com