Confiabilidad y análisis factorial de la versión en español del Indice de Calidad De Sueño de Pittsburgh en pacientes psiquiátricos / Reliability and factorial analysis of the Spanish version of the Pittsburg Sleep Quality Index among psychiatric patients

Introducción: El Índice de Calidad de Sueño de Pittsburgh (ICSP) se ha convertido en un instrumento estándar para la medición de la calidad del sueño. No obstante, no se ha evaluado su estructura factorial y, además, la existencia de dos versiones en castellano ha puesto de manifiesto la necesidad de realizar adaptaciones. El objetivo de este estudio fue evaluar la confiabilidad y composición factorial del ICSP. Material y métodos: Ochenta y siete pacientes psiquiátricos y 48 sujetos control se sometieron a una entrevista psiquiátrica y completaron el ICSP. Se estimó la consistencia interna y composición factorial del ICSP, se compararon las calificaciones por grupo y sexo, y se estimó la relación con la edad. Los grupos no difirieron significativamente en edad y sexo. Resultados: El ICSP obtuvo un coeficiente de confiabilidad satisfactorio (0.78) y coeficientes de correlación significativos (0.53 a 0.77) entre los componentes y la suma total, quedando conformado por dos factores: calidad de sueño per se y duración del sueño. Los pacientes presentaron calificaciones más altas que los sujetos control, tanto en la suma total como en los componentes, excepto en la duración del sueño. No se identificaron diferencias por sexo, ni relación significativa con la edad. Conclusiones: Estos resultados indican que el ICSP es un instrumento confiable para la evaluación de la calidad del sueño en población mexicana.

INTRODUCTION: The Pittsburgh Sleep Quality Index (PSQI) has become a standard instrument to measure sleep quality. However, its factor structure has not been fully explored and the available Spanish versions have revealed the need to make adaptations. Our objective was to assess the factor structure and reliability of the PSQI. METHODS: Eighty seven psychiatric patients without treatment and 48 control subjects underwent psychiatric structured assessment and completed the PSQI. Internal consistency and factor structure of PSQI was measured and the scale scores were compared between groups and by gender. Association with age was also calculated. RESULTS: There were no significant differences between psychiatric patients and control subjects on age and gender. Subjects had no problem understanding and answering the questions in the instrument. The PSQI displayed a satisfactory reliability coefficient (0.78) and component-total score correlations were all significant (0.53-0.77). The PSQI showed two main factors: sleep duration and sleep quality. Patients obtained significantly higher scores than controls, in both the global and the component scores, with the exception of sleep duration. The PSQI scores were not significantly different between males and females and were not associated with age. CONCLUSIONS: The PSQI is a reliable instrument to measure sleep quality in Mexican subjects.

Estudio de validación de la traducción al español de la Escala Atenas de Insomnio

resumen está disponible en el texto completo

Summary Introduction Insomnia affects 30% of population; 10% suffers from chronic insomnia. Quantitative approaches have predominated in the clinical assessment of insomnia; however, the importance of evaluating qualitative aspects has been outlined in the last decade. This has been reflected in diagnostic criteria of classification systems. According to DSM-IV-TR, ICD-10, and ICSD, the diagnosis of insomnia does not require the objective quantification of reduced sleep. For many years, sleep diaries/logs have been used in the measurement of insomnia, and although they contemplate the individual's subjective perception of sleep, they assess only sleep quantity. In addition to the sleep diaries/logs, several instruments for the measurement of sleep problems have been developed on the basis of different clinical and/or research needs; the Pittsburgh Sleep Quality Index (PSQI), the Leeds Sleep Questionnaire, and the St. Mary's Hospital Sleep Questionnaire are some of them. Recently, the Athens Insomnia Scale (AIS) has been presented. In contrast with previous instruments, AIS is based on the International Classification of Diseases (ICD-10) diagnostic criteria for insomnia. This paper presents the results of a translation into Spanish and the validity study of the AIS with a sample of Mexican population. Method The AIS is a self-rating instrument of eight items. In its instructions, the requisites for sleep problems frequency and duration are established, and correspond to criterion B of insomnia (ICD-10). Nevertheless, the time period of study can be modified to adjust it to research and/or clinical interests. The first four items of AIS asses sleep problems from a quantitative point of view, and the fifth item assesses sleep quality. These five items correspond to criterion A. The last three items evaluate the impact of insomnia during the day (criterion C). A simplified version, consisting of the first five items has been proposed by the authors. Each item can be rated in a 0-3 scale, where cero means the lack of problems and three the most severe condition. Total score is obtained from the sum of scores on eight items (range 0-24). The guidelines of the World Health Organization (WHO) were followed with the aim of securing the equivalence between Spanish and English versions of the AIS. In this way, a bilingual group of experts revised the conceptual structure of the original scale and approved its susceptibility of translation. The instrument was translated into Spanish and this text was examined by the bilingual group and a monolingual group. Minor corrections were suggested by the monolingual group. Then, the Spanish text was back translated into English by another bilingual expert. After revising this back-translated text, the bilingual group considered that the Spanish version was equivalent to the original. The sample consisted of a control group (n=146) of high school, undergraduate, and postgraduate students, and a clinical group of psychiatric outpatients (n=48) and inpatients (n=51). After giving their informed consent to participate, all subjects completed the AIS (eight items) with a modification in the time period of assessment. Internal reliability coefficient, total-item correlations, and differences in scores for gender and group were calculated. Also, the AIS was subjected to factorial analysis. Results The sample was composed by young adults with a slightly larger proportion of women (57%) than men. Psychiatric patients showed significantly higher scores than control subjects, and obtained higher scores than men, but this difference was not significant. Age showed a significant but weak positive relation with AIS scores. The AIS showed a high internal reliability in the whole sample (Cronbach's alpha=0.90). The lowest coefficient (0.77) was observed in the control group and the highest one (0.93) in psychiatric outpatients. Total-item correlations ranged from moderate to high; again, control patients obtained the lowest and psychiatric patients the highest. In the factorial analysis, the eight items emerged as a single component with a high percentage of explained variance (59.5%) and item-item correlations ranged from 0.38 to 0.75. Discussion These findings showed that AIS (Spanish text) is a useful instrument in the assessment of insomnia. Its brief and simple format let the clinician and/or researcher have a numerical index about sleep problems in just a few minutes. Another of the AIS's attributes is the possibility of changing the time period of assessment. This contrasts with some other instruments which require longer times for rating and scoring. Besides, some questionnaires are less flexible and are designed to asses predetermined time periods. Internal reliability obtained with the whole sample and with each group is almost identical to that obtained in the original report with a translation into Greek. This finding supports the efficiency of WHO's guidelines for translation of instruments, documents the cultural stability of certain constructs, and facilitates the integration of data from different investigations. Even when several instruments are available for the evaluation of sleep problems, the PSQI is probably the most used, and as far as we know the only one which translation into Spanish has been subjected to a validity study. In comparison with AIS, the Spanish versions of PSQI have shown a less stable internal consistency. Furthermore, total-item and item-item correlations have been less satisfactory for the PSQI than for AIS. The high degree of internal homogeneity of AIS is also supported by the results of factorial analysis where the eight items emerge as a single component, and total-item correlations ranged from moderate to high. As an additional evidence for AIS validity, women and psychiatric patients showed higher scores, and agerelated positively, but weakly, with AIS scores. Recently, the results of a diagnostic validity study with AIS have been published and a cutoff score of six has shown to be the best balance between sensitivity and specificity for a correct case identification. Future research should focus on the relation between AIS scores, age, and gender; the applicability of AIS in samples of patients with specific sleep disorders or different psychiatric disorders, and the AIS sensitivity to change (e.g., with or without pharmacological and/or no pharmacological interventions). In conclusion, the AIS is a brief and reliable instrument for the measurement of insomnia in clinical practice and sleep research.