Type 2 diabetes can affect—and be affected by—far more than blood glucose. Kidney function, blood pressure, cholesterol, sleep, food access, mental health, stress, inflammation, weight, medications, and cardiovascular disease often intersect. The newer term for this connected landscape is cardiovascular-kidney-metabolic syndrome, or CKM syndrome.

A new study uses machine learning to build a “CKM aging index,” designed to identify people with a higher risk of poor cardiovascular and kidney-related outcomes. The work is promising because it attempts to see the whole connected picture. It also needs careful interpretation.

A risk score can help researchers and clinicians ask better questions. It cannot tell a person their destiny, calculate a literal “true age,” or replace routine screening and individualized care.

What the researchers built

The PLOS Medicine study analyzed data from 6,896 U.S. adults in the National Health and Nutrition Examination Survey, or NHANES, from 2005 through 2018. The researchers evaluated more than 100 candidate survival models and selected a simpler Cox-model approach to create their cardiovascular-kidney-metabolic aging index, called CKMAI.

The score uses 18 variables that span several domains. They include age, sex, diabetes status, kidney measures such as eGFR and urine albumin-to-creatinine ratio, inflammatory or nutritional markers, smoking, and social and psychosocial factors such as income-related measures, education, marital status, and depressive symptoms.

The researchers then tested whether CKMAI predicted all-cause mortality, cardiovascular mortality, and a high-risk CKM category better than the comparison aging measures they selected. In the source dataset, the new index performed well on those statistical measures. The authors also performed validation across different time periods within NHANES and a preliminary external validation in a hospital-based Chinese cohort of 261 people.

That is meaningful work. Many risk tools focus on a narrow set of laboratory values. This score makes visible the idea that biology, daily conditions, and social context can coexist in a person’s health record.

Prediction is not the same as cause

The study is retrospective and observational. It looks for patterns in data that are associated with later outcomes. It cannot prove that a high score causes death, that a single biomarker causes kidney disease, or that changing one component will automatically change a person’s prognosis.

For example, the study found that depression statistically mediated part of the association between CKMAI and several outcomes. This is a prompt to take mental health seriously. It is not proof that depression alone causes cardiovascular-kidney-metabolic progression, and it should never be used to blame a person for their health.

Likewise, a risk score can identify higher-risk patterns without revealing the best next action for every individual. An elevated score might reflect age, an existing disease, social disadvantage, food insecurity, access barriers, sleep disruption, medication complexity, or several factors at once. Clinical care requires understanding that story, not simply labeling someone “older” or “high risk.”

The new score is not ready to replace routine care

The researchers are clear about important limitations. The primary analysis came from one U.S. data source. The external cohort was small and did not have long-term mortality follow-up. More validation is needed in larger and more diverse populations.

Just as important, the study did not test whether calculating CKMAI improves health outcomes. It did not randomize clinics to use the score, show that people received better care because of it, or prove that it changes the risk of heart attack, kidney failure, or death.

Those steps matter before a research score becomes routine clinical practice. A tool can have excellent statistical performance and still create problems if it is difficult to explain, used without patient consent, dependent on missing or biased data, or allowed to substitute for clinical judgment.

The score also includes social and demographic variables. These can help illuminate unequal risks created by unequal conditions. But they must be used with transparency and fairness. A social-risk marker should lead to more support—such as help with food access, transportation, medication affordability, or mental-health care—not less opportunity or more stigma.

What current CKM guidance already supports

The 2026 AHA/ACC/ADA/ASN clinical guideline describes CKM syndrome as the interconnection of metabolic risk factors, chronic kidney disease, and cardiovascular disease. The guideline focuses on earlier screening, risk-factor assessment, prevention, and coordinated care. Its public clinical summary highlights established approaches such as the PREVENT cardiovascular-risk equations; it does not recommend CKMAI as a routine standard.

The guideline’s broader message is more important than any one new algorithm. Blood pressure, glucose, cholesterol, kidney function, weight, physical activity, food patterns, sleep, tobacco exposure, mental health, and social conditions all deserve attention. Evidence-based medications and metabolic surgery may be appropriate for some people. Lifestyle support and social resources are not alternatives to medical treatment. They are part of comprehensive prevention and care.

Kidney screening remains especially important in type 2 diabetes. The ADA 2026 Standards recommend checking both eGFR and urine albumin-to-creatinine ratio at least annually in all people with type 2 diabetes. The most useful risk assessment is often not a novel score. It is making sure the fundamental monitoring and follow-up are actually happening.

A constructive way to use the “whole picture” idea

The best takeaway from CKMAI is not “find your biological age.” It is “ask whether care is seeing the whole picture.”

For a person with type 2 diabetes, that might mean asking whether kidney tests are current, whether blood pressure and cholesterol plans are working, whether sleep or stress is getting in the way, whether medication costs are manageable, whether food choices are constrained by access, or whether depression and anxiety deserve more attention.

For clinicians and health systems, it means treating social and mental-health information as part of high-quality care rather than as an unrelated footnote. It also means being careful with artificial intelligence and risk tools: validate them, explain them, measure whether they help, and ensure they do not deepen inequity.

At Diabetes Reversal Group, we believe this whole-person view is essential. Food, movement, sleep, stress support, relationships, access to care, regular kidney and cardiovascular monitoring, and clinician-directed medication all work together. Some people may pursue type 2 diabetes remission with sustained changes and medical follow-up. Remission is not a cure, and it is never the only measure of success.

A risk score should not make anyone feel defined by a number. At its best, it can remind us to look more closely, listen more carefully, and offer the right support earlier.

By: Dr. Jeffrey Hockings, CEO / Founder & Kristine Burke, MD, Chief Medical Officer

This article is educational and does not replace individualized medical care. Do not start, stop, or change diabetes, kidney, heart, or mental-health medication based on an online risk score, a “biological age” calculator, or a research result without guidance from the clinician managing your care.