Risk prediction models show 0.83 discriminatory performance for identifying malnutrition in dialysis patientsPredicting malnutrition risks for patients on dialysis

Renal failure Published June 19, 2026 Study authors: Liu Mengyao, Wu Yan, Ye Fen, Liu Wenting, Deng Xu, Tang Yang, Deng Lili PubMed ↗ DOI ↗ Editorial oversight: Dr. Amelia Tan, PhD · Internal Medicine & Chronic Disease

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway

Note that while models show 0.83 discriminatory performance, high risk of bias limits current clinical application.

This meta-analysis evaluates the prevalence of malnutrition and the performance of various risk prediction models in dialysis patients across 12 studies. The analysis identified a pooled prevalence of 41% for malnutrition among these patients. Furthermore, it identified 8 statistically significant predictors: age, serum calcium, Kt/V, triglycerides, sex, vitamin D, NT-proBNP, and comorbid diabetes.

The meta-analysis assessed the discriminatory performance of nine internally validated models, which yielded a pooled effect size of 0.83. While these results suggest some predictive capability, the authors note significant limitations regarding the quality of the evidence. All included models were rated at high risk of bias, and the analysis was hampered by inappropriate data sources and poor reporting.

Due to these methodological limitations and the high risk of bias in the source studies, the clinical applicability of existing models is constrained. The findings highlight a critical need for standardized, externally validated models rather than relying on currently available internal models. Clinical application should be interpreted with caution until higher-quality evidence is available.

How this fits prior evidence

This meta-analysis addresses a gap in identifying specific predictors and model performance for malnutrition in dialysis patients. It complements prior findings regarding machine learning models for malnutrition risk prediction in elderly ICU trauma patients, which noted that tree-based ensemble models may improve prediction accuracy. Unlike the ICU study, this analysis highlights significant methodological limitations and high risk of bias in current models used for dialysis populations.

Living with kidney disease often means undergoing regular dialysis. For many of these patients, a hidden but serious struggle is malnutrition. A large review of 12 studies found that about 41% of dialysis patients in China suffer from this condition.

Researchers identified several key factors that help predict who might struggle the most with nutrition. These include age, sex, and specific health markers like vitamin D levels, serum calcium, and triglycerides. Other important indicators included a marker called NT-proBNP and having a history of diabetes.

While the models used to track these risks showed strong performance in this review, there is a catch. The researchers noted that many of the original studies had a high risk of bias or poor reporting. Because of these limitations, doctors should be cautious about using these specific tools for clinical decisions until more standardized and validated models are available.

What this means for you:

About 41% of dialysis patients face malnutrition, with factors like age and vitamin D levels helping predict risk.

Common questions

How common is malnutrition among people on dialysis?

The study found a pooled prevalence of 41% for malnutrition among dialysis patients in China. This means nearly half of the patients in these studies were identified as having nutritional issues.

What specific factors help predict if a patient will be malnourished?

The research identified eight significant predictors: age, serum calcium, Kt/V, triglycerides, sex, vitamin D, NT-proBNP, and comorbid diabetes. These factors can help build models to identify patients at risk.

Are these prediction models ready for use in clinics?

The study notes that the models used had a high risk of bias and some had poor reporting or inappropriate data sources. Because of these methodological limitations, the results may not be directly applicable to clinical practice yet.

Study Details

Study typeMeta analysis

EvidenceLevel 1

PublishedDec 2026

PMID42315969

View Original Abstract ↓

Although multiple risk prediction models have been developed to identify malnutrition in dialysis patients, their quality and performance remain unclear, limiting their practicality in current clinical practice and future research. Therefore, we conducted a systematic review and meta-analysis to evaluate these models. Searches were conducted in PubMed, Embase, Web of Science, The Cochrane Library, CINAHL, SinoMed, CNKI, Wanfang, and VIP Database from inception to January 26, 2026. Two investigators independently screened the literature, extracted data, and assessed quality using the Prediction model Risk of Bias Assessment Tool (PROBAST). Meta-analyses of the prevalence of malnutrition, common predictors and model performance were performed using Stata 18.0 and R 4.5.1. A total of 12 eligible studies conducted in China were included, and the pooled prevalence of malnutrition in dialysis patients was 41%. Meta-analysis identified age, serum calcium, Kt/V, triglycerides, sex, vitamin D, NT-proBNP, and comorbid diabetes as statistically significant predictors. The pooled effect of the nine internal validated models was 0.83, indicating good discriminatory performance. However, all included models were rated at high risk of bias, primarily due to inappropriate data sources and poor reporting of the analysis. The current analysis reveals a high prevalence of malnutrition among dialysis patients. Eight significant predictors were identified, guiding future selection for constructing predictive models of malnutrition risk in this population. Although existing models demonstrate adequate discriminatory performance, their methodological limitations constrain clinical applicability. Future studies should prioritize the development of standardized, externally validated models to enable early identification and intervention, thereby improving outcomes in this vulnerable group.

Risk prediction models show 0.83 discriminatory performance for identifying malnutrition in dialysis patientsPredicting malnutrition risks for patients on dialysis

How this fits prior evidence

Common questions

Study Details

Systematic review confirms clip anastomosis improves one and two year primary patency rates for arteriovenous fistula creation compared to suture methods

Vascular Clips Outperform Sutures for Dialysis Access Longevity

Clinical research that matters. Delivered to your inbox.

Risk prediction models show 0.83 discriminatory performance for identifying malnutrition in dialysis patientsPredicting malnutrition risks for patients on dialysis

How this fits prior evidence

Common questions

More on Malnutrition

Study Details

Systematic review confirms clip anastomosis improves one and two year primary patency rates for arteriovenous fistula creation compared to suture methods

Vascular Clips Outperform Sutures for Dialysis Access Longevity

Clinical research that matters. Delivered to your inbox.

Related in Nephrology

From Other Specialties