**Background:** Chronic kidney disease (CKD) affects approximately 13% of the US population. Dietary guidelines for CKD have traditionally focused on restricting individual nutrients (e.g., sodium, potassium, phosphorus) rather than promoting whole foods like fruits and vegetables (F&Vs). F&Vs contain beneficial phytochemicals with antioxidant and anti-inflammatory properties, but patients with CKD may limit them due to concerns about dietary potassium. This study aimed to characterize patterns of F&V intake in US adults with and without CKD across three time periods.
**Methods:** The study used 24-hour dietary recall data from three NHANES cohorts: NHANES III (1988–1994), Continuous NHANES 2003–2010, and Continuous NHANES 2011–2018. CKD was defined as eGFR <60 mL/min/1.73 m² (using the 2009 CKD-EPI equation) or albuminuria (urine albumin-to-creatinine ratio >30 mg/g). F&Vs were categorized by processing level (unprocessed, minimally/processed, ultra-processed) following the NOVA classification system and by phytochemical content (glucosinolate-rich, carotenoid-rich, polyphenol-rich, starchy vegetables). Pairwise grouping yielded 12 potential categories, collapsed to 10 final categories for latent class analysis (LCA). LCA was performed separately for each cohort, with 4-class models selected based on Bayesian information criterion and clinical interpretability. Associations between CKD and F&V patterns were assessed using weighted multinomial logistic regression, adjusting for age, sex, race, waist circumference, diabetes, and hypertension.
**Key Results:** Across all cohorts, 13.6% (NHANES III) to 15.2% (Continuous NHANES 2011–2018) of the population had CKD. Participants with CKD were older (mean age 58.5–60.3 years vs. 41.0–44.6 years in non-CKD), more likely to be female and Black, and had higher BMI, hypertension, and diabetes prevalence. Four consistent F&V patterns emerged: Overall Low Intake, High Unprocessed, High Ultra-Processed, and Moderate Processed. The Overall Low Intake pattern was the most prevalent in all cohorts and CKD groups. Classification diagnostics showed mean posterior probabilities ranging from 0.60 ± 0.13 to 0.84 ± 0.18 across patterns and cohorts, with entropy of 0.59 (NHANES III) and 0.55 (Continuous NHANES). Dietary fiber was significantly lowest in the Overall Low Intake pattern (P < .0001), while carbohydrate and total energy were highest in the High Ultra-Processed pattern (P < .0001). Serum carotenoids (α-carotene, β-carotene, β-cryptoxanthin, lutein/zeaxanthin) were lower in the Overall Low Intake and High Ultra-Processed patterns (P < .0001). After adjustment for demographics and health conditions, participants with CKD were more likely to be classified in the Overall Low Intake pattern. The overall association was significant in NHANES III (P = .05) and Continuous NHANES 2003–2010 (P = .005), but not in Continuous NHANES 2011–2018 (P = .4). Sensitivity analyses using 1-step modeling yielded similar results.
**Clinical Implications:** This study provides evidence that low F&V consumption is more common among adults with CKD, even after adjusting for demographic and clinical factors. The findings challenge the traditional nutrient-restrictive approach to CKD dietary management and support a shift toward whole-food-based recommendations. However, because the data are cross-sectional, it remains unclear whether low F&V intake precedes CKD development or is a consequence of dietary modifications made after diagnosis (e.g., potassium restriction). The lack of statistical significance in the most recent cohort (2011–2018) may reflect a higher proportion of early-stage CKD (albuminuria-only) patients. Longitudinal studies are needed to clarify the direction of this association and to inform evidence-based dietary guidelines for CKD patients.