The study will leverage the established multi-stage cohort resources at Huashan Hospital, covering the entire spectrum from early-stage CKD to dialysis, including a comorbidity real-world database, a prospective CKD patient cohort, and a maintenance hemodialysis patient real-world cohort. It will also collaborate with the Shanghai Fifth People's Hospital (community-based CKD cohort, n>2,000) and the Jing'an District Central Hospital (end-stage renal disease cohort, n>400) as external validation platforms. Using epidemiological and big data-driven approaches, the study aims to elucidate the mechanisms underlying disease progression and complications, construct intelligent intervention models, and validate their performance in the external cohorts, ultimately facilitating the optimization and broader application of comprehensive CKD management strategies.

This study is a multi-center, bidirectional cohort study. The research data are primarily derived from the multidisciplinary clinical cohort of chronic kidney disease (CKD) established by the Department of Nephrology at Huashan Hospital, Fudan University, and are supplemented by external validation platforms at Shanghai Fifth People's Hospital and Shanghai Jing'an District Central Hospital.

The study will systematically integrate and utilize the multidimensional cohort resources already established at Huashan Hospital, covering the entire spectrum from early-stage CKD to dialysis, including a real-world database of comorbid populations, a prospective cohort of CKD patients, and a real-world cohort of maintenance hemodialysis patients. Using epidemiological and big data-driven computational approaches, the study will focus on exploring comorbidity patterns associated with CKD and establishing an early classification and early warning system; it will also investigate the mechanisms underlying the progression of kidney disease and related complications such as cardiovascular disease, nutritional and metabolic abnormalities, and osteoporosis, and develop intelligent intervention models. Based on these findings, the study will propose the concept of metabolic balance and accordingly establish precision nutrition and drug evaluation protocols.

To ensure the generalizability and reliability of the research outcomes, this study will conduct external validation and effectiveness evaluation of the previously constructed comorbidity patterns, early warning system, and management protocols in the community-based CKD cohort (over 2,000 participants) at Shanghai Fifth People's Hospital and the end-stage renal disease cohort (over 400 participants) at Jing'an District Central Hospital. This will facilitate the widespread application and continuous optimization of the comprehensive CKD management program.

Inclusion Criteria:

Prospective:

• Patients who have documented outpatient or inpatient records at Huashan Hospital, Shanghai Fifth People's Hospital, or Jing'an District Central Hospital during the period from September 2018 to September 2025.
• Aged 18-90 years.
• Diagnosed with one or more of the 48 pre-specified common chronic diseases according to ICD-10 codes.
• Patients diagnosed with end-stage renal disease (CKD stage 5) and receiving long-term maintenance hemodialysis, with a dialysis vintage of at least 3 months.
• Have relatively complete hemodialysis records available at the participating center.

Retrospective:

• Patients who meet the diagnostic criteria for chronic kidney disease (CKD) stages 1-5 (non-dialysis) as defined by the KDIGO guidelines, with a disease duration of at least 3 months.
• Aged 18-80 years, with normal cognitive ability to independently, or with the assistance of a caregiver, complete dietary diaries and quantitative assessments, and willing to provide written informed consent voluntarily.
• Able to comply with regular outpatient follow-up visits and willing to record a 3-day dietary diary (including both weekdays and weekends) as required.

Exclusion Criteria:

Prospective:

• Patients whose dialysis-related data and key clinical variables (e.g., creatinine, proteinuria, and other outcome-related indicators) are missing or severely illogical, and which are deemed unusable for research purposes upon evaluation.
• Patients with poor compliance during dialysis, or those receiving palliative dialysis.
• Other conditions that the investigators consider inappropriate.

Retrospective:

• Patients who have experienced acute kidney injury within 3 months prior to enrollment, or who are in an acute stress state such as acute infection, myocardial infarction, or other acute conditions at the time of screening.
• Patients with severe cognitive impairment, mental illness, or insurmountable communication barriers that prevent accurate provision of metabolic assessment information.
• Pregnant or lactating women; patients currently participating in other drug interventional clinical trials that may interfere with the outcome assessment of this study.

The study data originate from the real-world comorbidity database (covering 48 common chronic diseases) built from Huashan Hospital's inpatient EMR system, the real-world maintenance hemodialysis cohort based on China's first hemodialysis information management system at Huashan, the prospective CKD cohort from Huashan's integrated nephrology clinic, as well as the community CKD cohort from Shanghai Fifth People's Hospital and the ESRD cohort from Jing'an District Central Hospital for external validation.