Determination of the psychometric properties of the bace (Barriers to access to care evaluation) scale's Turkish translation

Introduction and aim: Young people who have started university may have difficulty in adapting to university life due to starting a new life, characteristics of developmental period, developmental tasks brought about during this period, inclusion in a new education system and financial difficulties. Young people experience many changes in terms of where they live, their educational practices and their social lives. Some of these students are perceived to be anxious, by some while these changes are more easily perceived and accommodated. However, there is a harmonization process in which whatever level of readiness they will have, they will absolutely live. In this process, it is known that the students who know and use the possibilities of the school and the nearby environment, show a superior development. Although it is known that the activities that can be organized within the orientation services to be offered to the students will facilitate the adaptation process of the students, such practices are not yet at the desired level in our country. In the context of psychological help-seeking, there are different problems for each individual. These are the factors that result from the lack of identification and assessment of the problem, social relationships, demographic factors, social support and services provided. A number of studies have been made in the world and in our country to determine the barriers to access to mental health services. The BACE scale is also a scale developed for this purpose and has not been used in our country to this day. The BACE scale was developed in the Health Services and Population Research Department of the Institute of Psychiatry, King's College, in London, England. The BACE scale is a 30-item self-report instrument conceived to evaluate barriers to access to mental health care. Respondents should indicate whether each item/ issue has ever stopped or delayed or discouraged him/her from getting or continuing to get professional care for a mental problem, by checking one of four possible answers: not at all (0), a little (1), quite a lot (2), or a lot (3). Therefore, in the present study, we aimed to determine of properties of the BACE Scale's Turkish translation. Materials and methods: Sample consisted of randomly selected 662 students from The Faculty of Science and Letters and The Faculty of Medicine at Adnan Menderes University who studied in 2016-2017 academic year; however, 47 questionnaires had to be excluded from the statistical evaluation. Data were gathered using the Turkish translation of the BACE Scale (version-III) and form of socio-demographic data. The study was designed as three-step psychometric validation study. The BACE scale was adapted to Turkish language. At this stage, the original scale was translated into Turkish by three researchers in accordance with WHO principles. Turkish text was later translated back into English, and its compatibility with the original language of the translation was determined. Personal Information Form that consists of 13 items was also added in questionnaire form, and we determined the final questionnaire form. In the first stage of the study, final questionnaire form was applied to 662 (47 forms were excluded) students. One month after the first application, questionnaire form was re-applied to 225 students in the previous surveyed classes, in order to determine the reliability of the test-retest of the our scale. Based on the results of the survey, the psychometric properties of the Turkish translation of the BACE scale were determined. Results: In our study, 615 students' questionnaire form was taken into statistical evaluation, and avarage age of pariticipants was 22,0±2,5. 321 participants (%52,2) were female and 601 (%97,7) were single. The youngest participant was 18 years old and the oldest was 36 years old. 319 (51.9%) of the participants were in the Faculty of Medicine and 296 (48.1%) participants were in the Faculty of Science and Letters. The average monthly income of 341 students (55.4%) was between 1432-4664 TL. 306 (49.8%) participants were smoking and 297 (48.3%) were drinking alcohol. The BACE scale consisted of 30 questions. However, because five questions (S14, S24, S27, S28, S29) were excluded from analysis, that associated with being married or having children, because of there were very few participants with this characteristics among our sample. In the reliability analysis for 25 items, alpha coefficient was found as 0,798; five items (S2, S5, S7, S22, S23) were subtracted, because the item-test correlation coefficient was below 0.20. Reliability analysis was repeated with the remaining 20 items; one more item (S25) was subtracted, because of it's item-test correlation coefficient was less than 0.20. The alpha coefficient calculated for the remaining 20-item scale was 0,828. Reliability analysis was performed once again with the remaining 19 items. The item-test (total) score correlations of the 19-item ranged from 0.20 to 0.61 and the alpha coefficient was 0.831. The scale was evaluated as reliable with the remaining 19 items. The Cronbach alpha (internal consistency) value of the whole scale was 0.83, the subscale of the stigma-related items was 0.76, and the subscale of the nonstigma-related items was 0.69 (p = 0,000). It can be said that the internal consistency of the scale is good, Although the sub-dimension of the non-stigma related items was being at the 0.70 border. The test-retest correlation, which determines the invariance of the scale to time, is 0.95; and subscale correlations were 0.94 and 0.93, respectively (p = 0,000). The strong correlation for the scale and its sub-dimensions indicates that the scale produces compatible scores at different times. Factor analysis was performed to show the scale's structure validity. Before the analysis, Kaiser-Meyer-Olkin (KMO) value was 0.839 and Bartlett's test was found significant at p = 0,000 level (χ2 = 2850). It was decided that the data were appropriate for factor analysis. The exploratory factor analysis showed that the scale was two-dimensional when the explained total variance values and the "scree plot" curve were taken into account. For this reason, factor analysis was repeated. The 13th and 30th items were removed from the analysis in turn, for the load values was below of 0.40 threshold. In the reconstructed analysis, there was no material showing an overlapping and below the threshold value of 0.40. The total variance explained by the two factors was 38.4%. Factor 1 was explaining for 27.2% of total variance, factor 2 for 11.1%. Factor load values of all items ranged from 0.41 to 0.75. The load values of all items were above the 0.40 value taken as the reference. However, as a result of subtracting thirteen items from the original 30-item scale, validity of the scale seems to be weak Conclusions: Studies to determine the barriers in front of access to mental health services have outshined in recent years in our country as well as in the whole world. In this study, we tried to gain the BACE scale's Turkish translation. The BACE Scale that we applied of 615 students has a good reliability but the valididty was not strong enough. However, the limitations associated with the group (university students) to which it is applied should not be overlooked. We believe that this scale, which has strengths to identify obstacles to access to mental health services, in particular stigma-related disabilities, needs to be reassessed in future studies with different groups. Since the psychometric properties of the original scale are quite good, we are optimistic that it can be reevaluated in future studies and new evidence of psychometric properties can be found. Keywords: University students, mental health, mental health services, access to mental health services, help seeking, barriers to access to mental heath services, The BACE Scale,