Uji Validitas dan Reliabilitas Cognitive Assessment Interview versi Indonesia

Abstract

Introduksi. Defisit kognitif merupakan prediktor konsisten disabilitas pasien skizofrenia. Salah satu instrumen penilaian fungsi kognitif pasien skizofrenia adalah Cognitive Assessment Interview. Tujuan penelitian ini untuk menguji validitas dan reliabilitas instrumen Cognitive Assessment Interview versi Bahasa Indonesia (CAI-Ina). Metode. Uji validitas dan reliabilitas meliputi uji validitas isi, uji validitas konstruk, uji reliabilitas interrater, dan uji reliabilitas konsistensi internal. Pengambilan data dilakukan satu kali pada 112 pasien skizofrenia dan informan di klinik rawat jalan psikiatri RS Jiwa Grhasia DIY. Analisis data menggunakan SPSS 16. Hasil. Hasil kajian ulang, penerjemahan, dan terjemahan balik CAI-Ina oleh pakar telah disetujui oleh Ventura sebagai penyusun instrumen asli. Semua item dalam instrumen CAI-Ina memiliki korelasi sedang-kuat dengan skor total CAI-Ina (r=0,568-0,720; p<0,001). Terdapat korelasi sangat kuat antara skor total penilai, pasien, dan informan (r=0,903-0,944; p<0,001). Analisis faktor mendapatkan dua komponen utama, yaitu komponen terkait memori dan tidak terkait memori yang keduanya dapat menjelaskan 58,6% keseluruhan konstruk instrumen CAI-Ina. Nilai Cronbach’s alpha pasien 0,789; informan 0,835; dan penilai 0,850. Hasil uji interrater koefisien Kappa seluruh item dan skor total 0,722-0,939 dan koefisien konkordansi Kendall W 0,834-0,985. Skor CAI-Ina memiliki korelasi kuat dengan skor GAF (r=-0,722-(-)0,808; p<0,001) ataupun kemampuan fungsi sosial (r=-0,684-(-)0,750; p<0,001). CAI-Ina memiliki validitas isi dan konstruk yang baik serta reliabilitas yang tinggi.

Introduction. Cognitive deficit is a consistent predictor for disabilities in patient with schizophrenia. One instrument that can be used to assess cognitive deficit in schizophrenia is Cognitive Assessment Interview. This study will evaluate the validity and reliability of CAI instrument in Indonesia languange version (CAI-Ina). Method. The validity and reliability evaluation includes content validity, construct validity, inter rater reliability, and internal consistency reliability. Data was collected one time from 112 patients and informants in the outpatient clinic in Grhasia Mental Hospital DIY. Data analysis was performed using SPSS 16. Result. Reassessment result, translation and back translation by an expert have been approved by Ventura as the original author. All items in the CAI have medium-strong correlation with the total score of CAI-Ina (r=0.568-0.720; p<0.001). There was a very strong correlation among the total score of the raters, patients, and informants (r=0.903-0.944; p<0.001). The two major components : memory related components and non-memory related components can explain 58.6% total instrument construct in the CAI-Ina. Cronbach’s alpha of patients was 0.789; of informants was 0.835; and of rater was 0.850. Kappa coefficients of all items and the total scores were 0.722- 0.939. The Kendall W concordance coefficients were 0.834 – 0.985. Score of CAI-Ina have a strong correlation with GAF score (r=-0.722-(-)0.808; p<0.001) and social functioning (r=-0.684-(-)0.750; p<0.001). CAI-Ina has a good content and constructs validity, also has a high reliability.