Abstract
Limited instruments are available to determine diet quality among US adults with chronic kidney disease (CKD). The purpose of this study was two-fold: (1) to develop a food frequency questionnaire, CKD SFFQ, for adults with CKD and (2) to validate the CKD SFFQ against two 24-h recalls in determining diet quality (DQ). A 57-item CKD SFFQ was developed through a content validation method. Adults with CKD (n = 46) completed the CKD SFFQ and 2-24-h recalls. Statistical analyses included descriptive statistics, frequencies, t-tests, Pearson correlations, and Bland-Altman plots. All data were analyzed using JMP SAS v15 with statistical significance detected at p < 0.05. Results showed no differences for the overall DQ (p = 0.11) and the nine whole-food components (p = 0.07 to p = 0.44) when comparing the CKD SFFQ to the 2-24-h recalls. Pearson correlation coefficients ranged from -0.39 (refined grains) to 0.60 (greens and beans). Bland-Altman plots showed overall good agreement and there was a systematic trend towards higher estimates with the CKD SFFQ, particularly for overall DQ, total proteins, and dairy. The majority of participants rarely or never consumed grains, fruits, vegetables, seafood, and plant proteins. The CKD SFFQ was demonstrated to be an acceptable method to determine DQ for adults with CKD.