Member's paper - Kidney Health Education and Research Group

Abstract

Validation of Patient Reported Outcomes Measurement Information System (PROMIS) Computer Adaptive Testing Depression Item Bank in Kidney Transplant Recipients

Shah, V., Rezaeishahreza, A., Lui, E., Aser, R., Bartlett, S., Li, M., Howell, D., Peipert, J., Novak, M., Mucsi, I.

2019

American Transplant Congress

Full Abstract
Purpose: To evaluate the reliability and validity of the PROMIS Depression computerized adaptive test (PROMIS-D CAT) in relation to a legacy measure in adult Kidney Transplant Recipients (KTR). Methods: A cross-sectional sample of adult outpatient KTR completed the PROMIS-D CAT and Patient Health Questionnaire (PHQ-9) on an electronic data capture platform. Socio–demographic and relevant clinical data was collected from medical records. Construct validity was assessed using Spearman’s Rho and known groups comparisons; test-retest reliability using intra-class correlation (ICC). Receiver operating characteristics (ROC) analysis was used to assess discrimination of PROMIS-D CAT using an established cut-off on the PHQ-9 (≥10) for moderate/severe depression. The PROMIS-D CAT cut-off for moderate/severe depression was identified using the Youden Index. Results: Mean (SD) age of the 156 participants was 50(16) years, mean (SD) CKD-EPI eGFR was 54(22), 67% was male and 64% Caucasian. Median (IQR) time since transplant was 8.5(3.4-13.8) years. Based on the PHQ-9, 17% reported moderate/severe depression. The PROMIS-D CAT and the PHQ-9 (r=0.59) were moderately correlated. Test-retest reliability was good (ICC=0.73). Mean (SD) PROMIS-D CAT scores were higher for patients with a Charlson Comorbidity Index ≥ 4 compared to <3 (51(7) vs 48(8); p=0.07) and patients with eGFR <30 compared to >60 (52(8) vs 49(9), p=0.07). Patients from minority groups had higher PROMIS-D CAT scores compared to Caucasians (51(8) vs 48(8); p<0.05). ROC analysis suggested excellent discrimination with an area under the curve of 0.91 (95% CI: 0.85-0.97). The Youden Index identified cut-off for moderate/severe depression for the PROMIS-D CAT was 54 (Sensitivity: 88%, specificity: 82%, positive predictive value: 0.5, negative predictive value: 0.97). Conclusion: These results provide new evidence of the validity and reliability of PROMIS-D CAT in KTR and support its use in research and care. In addition, a previously-calculated crosswalk between the PHQ-9 and PROMIS-D CAT allows for easy conversion among centers using the PHQ-9.

Member’s paper

Abstracts