Latar belakang: Pengkajian Paripurna Pasien Geriatri (P3G) bermanfaat dalam pelayanan pasien lanjut usia. Domain yang sering dinilai adalah status fungsional, disabilitas, status nutrisi, dan kognitif. Namun, pengerjaan P3G membutuhkan waktu yang lebih lama, sehingga dikembangkan bentuk singkat P3G, di antaranya Geriatric 8 (G8). Belum ada publikasi mengenai kesahihan, keandalan, dan performa diagnosis G8 pada populasi umum lansia di Indonesia. Tujuan: mengetahui kesahihan, keandalan, dan performa diagnostik G8. Metode: Penelitian ini merupakan penelitian potong lintang yang dilakukan di Poli Geriatri RSCM. Dilakukan pemeriksaan G8 dan P3G terhadap pasien yang memenuhi kriteria seleksi subjek penelitian, kemudian dilakukan uji kesahihan dengan mencari koefisien korelasi dan analisis kappa. Pasien dengan gangguan pada satu domain P3G dikatakan gangguan P3G, yaitu ADL  19, IADL  7, MoCA-INA  25, MNA < 24, atau timed up and go  10 detik. Hasil: terdapat 80 orang subjek penelitian dengan rerata usia 73,68 tahun, Interrater dan intrarater concordance masing-masing adalah 1 dan 0,904 (p<0,005). Interclass corelation coefficient berkisar antara 0,77 (0,412 – 0,913) sampai dengan 1 (1 – 1). Didapatkan nilai Cronbach 0,697. Titik potong acuan yang digunakan 14,25 menunjukkan sensitivitas 70,27 (58,82 – 80,34), spesifisitas 83,33 (35,88 – 99,58), dengan AUC 0,846 (p<0,005), IK95% 0,667- 1,0) Simpulan: G8 cukup sahih dan memiliki keandalan yang baik sebagai instrumen penapisan pasien rawat. Titik potong G8 yang disarankan adalah 14,5 sehingga pasien dengan skor lebih rendah disarankan untuk menjalani pemeriksaan P3G lengkap.

---

Background: Comprehensive geriatric assessment (CGA) has been proved to be beneficial for older adults care. Domains that usually assessed in CGA are functional status, disability, cognitive function, and nutrtion status. However, CGA takes more time to complete, hence shorter versions of CGA were developed, including Geriatric 8 (G8). G8 was developed to screen older adults with cancer who would benefit the complete CGA. There was no publication regarding validity, reliability, and diagnostic performance of G8 for general population of older adults in Indonesia. Objective: This study aimed to evaluate validity, reliability, and diagnostic performance of G8 in older adults. Methods: This is a cross-sectional study conducted in Geriatric Clinic of Cipto Mangunkusumo National Hospital. Both CGA and G8 were performed, concordance between these tests were analyzed to determine validity, reliability, and diagnostic performance of G8. Abnormal CGA is defined by at least one abnormal CGA domain is identified, i.e ADL  19, IADL  7, GDS  5, MoCA-INA  25, MNA < 24, or timed up and go  10. Commorbidities was assessed by CIRS-G. Results: We found strong inter-rater and intra-rater condordance (kappa=1 and kappa=0.904, p<0.005, respectively). Interclass Coefficient Corelation was ranged 0.77 (0.412 – 0.913) to 1 (1 – 1). We also found acceptable Cronbach of 0.697. For diagnostic performance, the sensitivity was 70.27 (58.82 – 80.34), specificity 83.33 (35.88 – 99.58), with AUC 0.846 (p<0.005), CI95% 0.667-1.0). Conclusion: G8 screening tool is valid and reliable to be used in older adults. G8 also demonstrated good diagnostic performance. We propose 14.5 as cut off point for older adults who need full form geriatric assessment.