2009, Number S1
Salud Mental 2009; 32 (S1)
Validación de constructo de la escala de autorreporte del Trastorno por Déficit de Atención con Hiperactividad (TDAH) en el adulto de la Organización Mundial de la Salud en población universitaria mexicana
Reyes ZE, Cárdenas GEM, García VKL, Aguilar ONC, Vázquez MJ, Díaz FA, Díaz FLA, Jaimes MA, Ortiz LS, Náfate LO, Gaspar BCE, Feria AM, Peña OF, Palacios CL
Language: Spanish
References: 20
Page: 69-75
PDF size: 169.59 Kb.
ABSTRACT
Over the last thirty years, there has been greater acceptance of the persistence of attention deficit/hiperactivity disorder (ADHD) into adulthood. Epidemiological data from open adult population show an ADHD prevalence at nearly 4%. ADHD evaluation in adults is a complex procedure that requires several things from the clinician: a) Symptoms should not be evaluated only as present or absent, since some symptoms may have been present during childhood and remitted since or may continue to be present but express differently in adulthood; b) Diagnostic criteria not designed for adults should not be used; c) The possible impact of the disorder in a patient’s life should always be considered, specially in areas such as substance abuse, social and personal relations or low academic achievement; d) Differential diagnostic is always necessary, since several disorders commonly present in adults have symptoms that may be similar to those of ADHD. Thus, in order to correctly establish an ADHD diagnostic, the evaluation procedure necessary is a difficult process in clinical population, and even more so in open population. Because of this, a valid screening instrument that correctly separates probable cases from non-cases would be useful for health and education professionals in daily practice since, with it, subjects that may require further evaluation (through semi-or highly-structured interviews) could be selected. The Adult ADHD Self-Report Scale Symptom Checklist (ASRS) was developed during the review of the World Health Organization Composite International Diagnostic Interview. The ASRS has two versions, a long one with 18 questions, each of which evaluates the frequency of one of the 18 symptoms according to DSM-IV criteria A for the disorder, and a short screening one made out from the first six ítems of the long version. These six ítems were selected after a logistic regression analysis in order to optimize the prediction of final clinical classification. Even though the short version may have certain advantages over the long one when used for evaluating wide populations, it was decided to use both versions on the present study.Objective: To establish the construct validity of the ASRS in a Mexican population of college students.
Methods: Five hundred-forty college students were evaluated (355 women; with a mean age of 23.34 years) from four different universities in Mexico City. Every subject answered the long version of the ASRS v.1.1. College authorities were contacted by the researchers and asked for their approval in order to invite students to participate in the study; once it was granted, students were contacted during the first ten minutes of a class. One of the researchers explained the goal of the study and asked students for their informed consent. Once the student granted consent, he was given the ASRS. Only five students (from the original 545 contacted) denied consent. Subjects that acceded and whose screening was positive were invited to participate in a second phase of the main study, where the diagnostic was confirmed by trained specialists. Statistical analysis: The internal consistency of the scale was tested by the Cronbahch’s alpha method. The ASRS structure was analyzed by a factorial analysis using the main components as a method of extraction and Varimax rotation. Only factors with an initial Eigen value higher than one were considered. An ítem was considered part of a factor only if it had a load higher than 0.4. Due to the fact that, according to past research, there should be differences in symptoms present in females versus those present in males, it was suggested that, once ítems were arranged into factors, there would be differences among males and females with ADHD on these factors. In order to test this hypothesis, males and females with positive screenings were compared on a factorial score calculated for each factor through the regression method. Considering that a student population was used for this work, authors decided to analyze, in a sub-sample of 125 psychology students, the possible relation between total and factorial scores of ASRS and academic achievement as measured by the average grade at the end of the semester. To test this second hypothesis, a Pearson’s correlation coefficient was calculated.
Results: Of the 540 subjects that consented to the study, 60 (11.1%) were eliminated from the analysis because they omitted answering one or more ítems. Reliability analysis of the scale showed that it has a good internal consistency, since Cronbach’s alpha was 0.88. It is noteworthy that no ítem unduly influenced the alpha coefficient since, when it was recalculated for the 18 total scores created by deleting one ítem at a time, there was no difference (range: 0.87-0.88). Factorial analysis found three factors that explained 49.26% of the total variance. Factor one contained eight ítems, and the one with the highest factorial load was: «How often do you have trouble wrapping up the final details of a project, once the challenging parts have been done?». Factor one was named «inattention factor». Factor two grouped five ítems; out of these, the one with the highest factorial load was: «When you’re in a conversation, how often do you find yourself finishing the sentences of the people you are talking to, before they can finish them themselves?». Factor two was named «impulsivity factor». The third and last factor was formed by four ítems; the one with the highest factorial load was: «How often do you feel restless or fidgety?». This was labeled «hyperactivity factor». The analysis by gender of ADHD symptoms performed on the three factorial scores calculated from the 121 positive screenings (48 males and 71 females) found statistically significant differences only in the impulsivity factor (t= 3.52, p= 0.00). This difference is explained by the greater number of symptoms of this factor reported by females. Finally, the correlation between ASRS and academic achievement found only one significant negative correlation between the factorial score of the inattention factor and the average grade of the student (r= -0.22, p= 0.02).
Conclusions: The psychometric characteristics of the Spanish version of the ASRS v.1.1 are adequate; because of this, its use as a tool to detect probable cases of ADHD in adult Mexican population seems to be well supported.
REFERENCES