The MetaboHealth score Predicts 10-Year All-Cause Mortality Risk

#1

This is another mortality risk NMR-based blood test (similar to the MVX test, that you can purchase today) that is being developed by an academic group in the Netherlands. It has not been made commercially available yet. I am posting the research studies that have been done on it in this thread.

14-Biomarker Blood Signature Predicts 10-Year Mortality Risk Better Than Standard Clinical Panels

A large-scale observational study of 44,168 human subjects identified a specific profile of 14 circulating metabolic biomarkers that accurately predicts all-cause mortality over a 5 to 10-year period. By utilizing high-throughput nuclear magnetic resonance metabolomics, researchers demonstrated that this single biomarker score outperforms conventional clinical risk factors like cholesterol and blood pressure across all age groups, offering a superior tool for assessing physiological frailty and longevity.

Traditional predictors of mortality such as systolic blood pressure and total cholesterol lose their predictive power in older populations. In elderly cohorts, these standard markers often show inverted associations with mortality due to a phenomenon known as mortality crossover, where low blood pressure or low cholesterol paradoxically signals failing health and impending death. Clinicians and researchers require better molecular tools that measure systemic physiological decline rather than singular disease states.

To solve this, researchers analyzed blood samples from over 44,000 individuals across 12 distinct cohorts spanning ages 18 to 109. Using a standardized nuclear magnetic resonance platform, the team screened 226 metabolic biomarkers and narrowed them down to 14 independent variables that robustly predict all-cause mortality. This mortality signature captures data across multiple physiological domains including lipoprotein metabolism, fatty acid balance, glycolysis, fluid balance, and systemic inflammation.

The 14 biomarkers include metabolic substrates and lipid particles where higher levels are protective, alongside inflammatory markers and metabolites where higher levels increase mortality risk. Protective markers include the ratio of polyunsaturated fatty acids to total fatty acids, total lipids in specific high-density and very low-density lipoproteins, albumin, and the amino acids histidine, leucine, and valine. Conversely, elevated glucose, lactate, phenylalanine, isoleucine, acetoacetate, and glycoprotein acetyls indicate higher mortality risk.

When combined into a single weighted score, this metabolic profile significantly improved mortality prediction accuracy compared to a standard clinical panel that included body mass index, blood pressure, cholesterol, smoking status, and existing disease. The predictive superiority of the biomarker panel was especially pronounced in individuals over the age of 60. The findings establish that a comprehensive snapshot of circulating metabolites provides a more accurate representation of biological age and survival probability than traditional physician screening panels.

Actionable Insights

For individuals optimizing for healthspan, this paper provides highly specific metabolic targets. The hazard ratios provided in the study allow us to extract the standardized effect size of each biomarker, translating directly to practical risk management.

Increasing the ratio of polyunsaturated fatty acids to total fatty acids yields a massive protective effect. For every standard deviation increase in this ratio, mortality risk drops by 22 percent (Hazard Ratio 0.78). Biohackers can act on this by prioritizing marine omega-3 fatty acids and minimizing saturated and trans fats to optimize cellular membrane composition.

Systemic inflammation is a primary driver of mortality. Glycoprotein acetyls, a marker of chronic inflammation, strongly predict death. A one standard deviation increase in this inflammatory marker increases mortality risk by 32 percent (Hazard Ratio 1.32). Interventions targeting chronic inflammation are non-negotiable for longevity.

Metabolic flexibility and glycemic control remain paramount. Elevated fasting glucose increases mortality risk by 16 percent per standard deviation, while elevated lactate increases risk by 6 percent.

The cumulative effect of these variables is profound. A one-unit increase in the combined 14-biomarker risk score results in a 173 percent increase in all-cause mortality risk (Hazard Ratio 2.73). Optimizing these specific metabolic hubs provides a significant, mathematically verifiable survival advantage.

Context and Source

Metabolic Vulnerability Index (MVX) Can Spot Who Dies First, Thirty Years Ahead of Time
#2

Blood Score for Death Risk Unmasks Hidden Inflammation and Lipid Failures in Aging

A new study investigates the biological mechanisms behind the MetaboHealth score, a 14-metabolite blood test highly predictive of 5-year mortality. By comparing individuals with extreme high and low scores across three large Dutch cohorts, researchers discovered that poor metabolic health is driven primarily by elevated pro-inflammatory cytokines like GDF15 and a severe collapse in proteins responsible for cholesterol transport. Twin data revealed that this mortality-linked score is moderately heritable, suggesting that lifestyle and pharmacological interventions targeting inflammation and lipid transport can meaningfully alter this trajectory of physiological frailty.

Biological age predictors accurately stratify mortality risk, but the underlying physiological drivers often remain opaque. The MetaboHealth score utilizes 14 specific metabolites to predict mortality with greater accuracy than conventional clinical variables. To determine the precise proteomic and inflammatory pathways captured by this score, researchers isolated 150 individuals representing the absolute high and low extremes from three cohorts: the Leiden Longevity Study, the Rotterdam Study, and the Netherlands Twin Register.

The proteomic analysis revealed a stark dichotomy between the high-risk and low-risk groups. A high MetaboHealth score, indicative of poor health and higher mortality risk, mapped heavily to systemic inflammation. The high-risk group exhibited significantly elevated levels of GDF15, IL6, and MIG (CXCL9). Furthermore, acute phase inflammatory proteins, specifically C-reactive protein (CRP), lipopolysaccharide binding protein (LBP), and haptoglobin (HPT), surged in the high-risk cohort.

Conversely, the data demonstrated a systemic failure in lipid metabolism among individuals with a poor MetaboHealth score. The unhealthiest participants showed a drastic reduction in proteins essential for reverse cholesterol transport and high-density lipoprotein remodeling, most notably APOA1, APOA2, and APOA4. This confirms that the physiological frailty captured by the MetaboHealth score is intertwined with how the body manages cholesterol clearance and immune homeostasis.

By analyzing monozygotic twins discordant for the MetaboHealth score, the researchers proved the score has a heritability of 0.4. The majority of the protein shifts, including 68 serum proteins, remained significantly different between genetically identical twins with opposing scores. This establishes that environmental, pharmacological, and lifestyle factors dominate this metabolic death-predictor, leaving ample room for targeted geroprotective interventions.

Actionable Insights

The MetaboHealth score is roughly 60 percent driven by non-genetic factors, making it a highly modifiable target for longevity interventions. The data indicates that suppressing systemic inflammation and optimizing lipid transport are the highest-yield strategies to improve this specific frailty metric.

The effect sizes observed in this study are massive. The high-risk group exhibited a 1.51 standard deviation increase in CRP and a 1.46 standard deviation decrease in APOA1. In practical terms, this translates to a Cohen’s d of approximately 1.5, meaning an individual in the high-risk group has a CRP level higher than 93 percent of the low-risk population. Similarly, GDF15 levels were elevated by 1.08 standard deviations in the unhealthy cohort.

Maintaining optimal lipid profiles through aggressive pharmaceutical management of ApoB and LDL, combined with rigorous cardiovascular conditioning to support lipid remodeling, directly counters the metabolic decay measured by this score. Routine tracking of high-sensitivity CRP is critical, as maintaining baseline inflammation near zero limits the age-related upward drift of senescence markers like GDF15 and MIG.

Context/Source

#3

They have now narrowed the test down to 10 key biological markers, from the original 14 markers.

A Blood Test for Biological Time: How a 10-Biomarker Panel Predicts Disease Onset and Responds to Lifestyle Changes

The Personal-MetaboHealth score is a novel, individualized metabolic biomarker panel based on ten clinically validated blood metabolites that predicts morbidity and mortality. Validated in the human Leiden Longevity Study, the score demonstrates that a one-unit increase corresponds to a 13.69-year delay in the onset of first cardiometabolic disease. Furthermore, data from the Growing Old Together study reveals that a 13-week lifestyle intervention of moderate calorie restriction and increased physical activity can significantly improve this score in older adults who possess unhealthy baseline metabolic profiles.

The transition from chronological age to biological age is the central focus of longevity medicine. Historically, molecular clocks and biomarker panels have been limited by two distinct problems. First, they are often scaled to a specific cohort, making it impossible to evaluate an individual patient in a clinical setting. Second, they often rely on experimental markers that lack regulatory approval. The Personal-MetaboHealth score addresses both limitations by using a reference population to allow individual scaling, while restricting the panel strictly to ten biomarkers that hold CE-mark clinical validation.

The research team evaluated this new scoring system using two distinct human datasets. The first was the Leiden Longevity Study, providing long-term prospective data on 2,404 middle-aged individuals with up to 22 years of mortality follow-up and 16 years of morbidity follow-up. The predictive strength of the test was substantial. The Personal-MetaboHealth score reliably predicted both all-cause mortality and the precise timing of first cardiometabolic disease onset. The data strongly suggests that metabolic signatures of aging precede clinical diagnosis by over a decade.

The second dataset utilized the Growing Old Together trial to determine if the biomarker is actionable. Biomarkers are only practically useful if they respond to behavioral or pharmacological interventions. In a 13-week trial involving older adults, participants underwent a 25 percent improvement in energy balance, split evenly between reduced caloric intake and increased physical activity. The intervention successfully improved the Personal-MetaboHealth scores for participants who started in an unhealthy state. Interestingly, participants who already possessed a healthy baseline score experienced minor fluctuations but generally maintained their healthy status, indicating that interventions yield the highest measurable molecular return on investment in those with existing metabolic dysfunction.

The practical outcome of this paper is the validation of a clinically approved, actionable blood panel that provides a highly specific timeline for cardiometabolic disease onset. This provides clinicians and individuals with a regulatory-approved tool to measure the direct efficacy of longevity protocols in real time.

Actionable Insights
For individuals seeking actionable pathways to improve their healthspan, the Personal-MetaboHealth panel isolates ten specific targets: the ratio of polyunsaturated to total fatty acids, glucose, albumin, glycoprotein acetyls, phenylalanine, isoleucine, leucine, valine, histidine, and lactate.

The magnitude of optimizing these markers is severe. The researchers calculated a standardized effect size demonstrating that every one-standard-deviation increase in the score reduces the yearly risk of developing a cardiometabolic disease by 26 percent (Hazard Ratio 0.74). In absolute terms, each one-unit increase in the Personal-MetaboHealth score delays the median onset of first cardiometabolic disease by exactly 13.69 years.

To achieve these improvements, the data confirms that aggressive pharmaceutical intervention is not strictly necessary for baseline correction. A protocol consisting of a 12.5 percent reduction in caloric intake coupled with a 12.5 percent increase in daily physical activity over just 90 days was sufficient to significantly reverse unhealthy scores (standardized beta improvement of 0.26). The primary driver of this improvement was the normalization of branched-chain amino acids, glucose, and inflammatory glycoproteins.

Context/Source

Biomarker Data (Effect Size Calculation) The biomarker effect sizes demonstrate high clinical relevance rather than mere statistical significance.

  • Mortality: A one-standard-deviation increase in the Personal-MetaboHealth score reduced all-cause mortality risk by 25 percent (Hazard Ratio = 0.75, 95% CI: 0.69 to 0.82).

  • Morbidity (Relative): A one-standard-deviation increase reduced the risk of first cardiometabolic disease onset by 26 percent (Hazard Ratio = 0.74, 95% CI: 0.67 to 0.81).

  • Morbidity (Absolute): Using Accelerated Failure Time modeling, a one-unit increase in the score generated a time ratio of 1.36. This equates to a 36 percent delay in disease onset, representing a median absolute delay of 13.69 years [Confidence: High].

  • Intervention Efficacy: In the 13-week lifestyle intervention, individuals with an unhealthy baseline score (< 0) achieved a statistically significant standardized score improvement of 0.26 (95% CI: 0.10 to 0.42).

Mechanistic Deep Dive The Personal-MetaboHealth panel heavily tracks the mechanistic targets of rapamycin (mTOR) and systemic inflammation. The inclusion of three branched-chain amino acids (leucine, isoleucine, and valine) acts as a direct proxy for mTORC1 complex activation. Chronically elevated branched-chain amino acids correlate with insulin resistance and the suppression of cellular autophagy.

Furthermore, the tracking of glycoprotein acetyls (GlycA) provides a high-resolution systemic view of inflammaging. GlycA is a composite marker for acute-phase proteins and reflects chronic innate immune activation, likely intersecting with the NF-kB and cGAS-STING pathways as tissues accumulate senescent cell burdens. The intervention’s success in lowering these specific amino acids and inflammatory markers strongly suggests that moderate caloric restriction paired with exercise effectively downregulates hyperactive mTOR and blunts systemic immune hyperactivation [Confidence: Medium]. The inclusion of lactate and glucose underscores the priority of mitochondrial bioenergetics, as age-related mitochondrial dysfunction forces a metabolic shift toward glycolysis.

#4

From Google Gemini, on Commercial Sources:

The best option we have today is the test we can purchase now: Metabolic Vulnerability Index (MVX)

The specific MetaboHealth score—the 14-biomarker NMR profile developed by the LUMC consortium—is not currently sold directly to consumers under that brand name by any major US commercial laboratory (such as Quest Diagnostics or Labcorp).

While the underlying proton nuclear magnetic resonance (1H-NMR) technology is commercially viable, its direct-to-consumer (DTC) footprint in the United States remains restricted. Here is the current commercial landscape and associated costs:

1. Nightingale Health (B2B and Research Only in the US)

Nightingale Health, the Finnish company whose NMR platform powers the MetaboHealth score, operates in the US but primarily focuses on B2B partnerships and academic research rather than a DTC storefront.

  • Availability: They offer the Nightingale Kit, a self-collected finger-prick blood test that analyzes the necessary metabolic biomarkers (amino acids, glycolysis markers, GlycA, detailed lipids). However, this is marketed to consumer health companies to white-label, or to research cohorts, rather than individual retail buyers.
  • International Price Benchmark: For context, in international markets where clinical partners offer the Nightingale Health Check directly to patients (such as ATA Medical in Singapore), the base screening costs approximately $90 USD. Upgraded packages that include full metabolic age breakdowns and associated blood values cost around $280 to $300 USD.

2. Comparable Commercial Alternatives in the USA

If you are looking for commercially accessible metabolic vulnerability scoring in the United States, there are closely related proxies available via DTC lab brokers:

  • Labcorp’s Metabolic Vulnerability Index (MVX): Labcorp recently validated its own blood-based prognostic score (0–100) called the Metabolic Vulnerability Index, which also predicts all-cause mortality and metabolic dysfunction (specifically liver decompensation and MASH), recently published in Nature Communications.
    • Cost: This test can be ordered through independent DTC lab brokers (such as OwnYourLabs) for approximately $40, utilizing Labcorp’s infrastructure.
#5

The 2026 Personal-MetaboHealth score relies on a refined panel of 10 proton nuclear magnetic resonance (1H-NMR) biomarkers that possess formal clinical validation (CE-marking).

To transition the algorithm from a cohort-level epidemiological tool to an individualized clinical diagnostic, researchers restricted the inputs to the following 10 markers, stratified across four physiological domains:

1. Branched-Chain and Aromatic Amino Acids

  • Leucine, Isoleucine, Valine: These three branched-chain amino acids (BCAAs) are critical nutrient sensors that directly regulate the mTOR (mechanistic target of rapamycin) signaling pathway. Elevated circulating BCAAs correlate strongly with mitochondrial dysfunction, insulin resistance, and impaired metabolic flexibility.

  • Phenylalanine & Histidine: Aromatic and essential amino acids where altered serum concentrations reflect shifts in systemic protein turnover and early hepatic stress.

2. Glycolysis and Bioenergetics

  • Glucose: The primary indicator of systemic glycemic control and insulin sensitivity.

  • Lactate: An end-product of anaerobic glycolysis. Chronically elevated baseline lactate is an indicator of compromised mitochondrial oxidative phosphorylation and poor overall bioenergetic efficiency.

  • Albumin: The most abundant circulating protein. Reduced levels serve as a proxy for hepatic synthetic dysfunction, systemic frailty, and increased biological age.

3. Systemic Inflammation

  • Glycoprotein Acetyls (GlycA): A highly stable composite marker of systemic inflammation that integrates the protein signals of multiple acute-phase reactants (e.g., alpha-1-acid glycoprotein, haptoglobin, alpha-1-antitrypsin). It serves as a superior proxy for chronic, low-grade inflammatory burden (“inflammaging”) compared to highly volatile markers like hs-CRP.

4. Lipid Metabolism

  • PUFA/FA Ratio: The ratio of polyunsaturated fatty acids to total fatty acids. Higher PUFA ratios are broadly cardioprotective and act as an indicator of favorable dietary fat composition and cellular membrane fluidity.

Biomarkers Excluded from the Original Panel

To achieve the CE-marked Personal-MetaboHealth iteration, the Leiden University Medical Center (LUMC) consortium deliberately excluded four variables that were present in the original 2019 14-marker mortality algorithm. These were dropped due to a lack of formal clinical validation on the Nightingale Health diagnostic platform:

  1. Acetoacetate: A ketone body.

  2. XXL-VLDL-L: Total lipids in chylomicrons and extremely large very-low-density lipoproteins.

  3. S-HDL-L: Total lipids in small high-density lipoproteins.

  4. VLDL-D: Mean diameter of VLDL particles.

Knowledge Gaps & Clinical Limitations

While the biomarker identities are public, the exact statistical weights applied to scale these 10 markers against the reference population remain partially obscured, complicating independent replication of the exact scoring model outside the LUMC consortium.

Furthermore, while the score reliably predicts cardiometabolic disease onset, causation remains unproven. Extensive clinical trial data is required to determine whether pharmacologically targeting specific markers within this panel (for example, utilizing mTOR inhibitors to suppress circulating BCAAs or senolytics to lower GlycA) directly extends organismal lifespan, or if such interventions merely “game” the biomarker score without altering fundamental biological aging.

#6

Comparing The Personal-MetaboHealth score to the Metabolic Vulnerability Index (MVX)

Biomarker-by-Biomarker Comparison

Both the Metabolic Vulnerability Index (MVX) and the Personal-MetaboHealth score utilize proton nuclear magnetic resonance (1H-NMR) metabolomics to assess physiological frailty, but they deploy distinct biomarker panels.

Overlapping Biomarkers (Present in both tests):

  • Valine, Leucine, Isoleucine: Branched-chain amino acids (BCAAs).

  • Glycoprotein Acetyls (GlycA): A composite marker of systemic inflammation.

Unique to MVX (6 total biomarkers):

  • Citrate: An intermediate of the tricarboxylic acid (TCA) cycle, reflecting mitochondrial energy status.

  • Small HDL Particles (S-HDL-P): A metric of reverse cholesterol transport and atherogenesis.

Unique to Personal-MetaboHealth (10 total biomarkers):

  • Glucose & Lactate: Direct markers of systemic glycemic control and anaerobic bioenergetics.

  • Albumin: A proxy for hepatic synthetic function and systemic resilience.

  • Phenylalanine & Histidine: Aromatic amino acids reflecting protein turnover and hepatic stress.

  • PUFA/FA Ratio: The proportion of polyunsaturated fatty acids to total fatty acids, indicating dietary lipid quality and membrane fluidity.

Similarities and Differences

Similarities:

Both panels recognize the prognostic superiority of GlycA over high-sensitivity C-reactive protein (hs-CRP) for tracking chronic, low-grade systemic inflammation (inflammaging), as GlycA is less susceptible to transient acute spikes. Furthermore, both algorithms heavily weight BCAAs. Because BCAAs are nutrient sensors that directly activate the mechanistic target of rapamycin (mTOR) pathway, elevated circulating levels consistently correlate with insulin resistance, mitochondrial dysfunction, and metabolic inflexibility.

Differences:

MVX is structurally divided into two sub-scores: the Inflammatory Vulnerability Index (IVX) and the Metabolic Malnutrition Index (MMX). It relies heavily on citrate and lipoprotein metrics while completely omitting direct glycemic markers. In contrast, Personal-MetaboHealth provides a broader physiological snapshot. It integrates glycolysis, hepatic function, and systemic dietary lipid profiles into its algorithm, shifting the focus from acute mortality prediction toward tracking mid-life metabolic interventions.

Summary of Comparative Research

Recent epidemiological studies have applied both algorithms concurrently in large biobanks to evaluate disease risk. For example, a 2026 prospective cohort study evaluated both the MVX and the MetaboHealth score to determine risk factors for incident age-related macular degeneration (AMD).

The broader literature establishes a clear functional divergence between the two indices. MVX functions as a highly sensitive, chronological age-independent predictor of near-term morbidity. It reliably identifies patients at risk for 5-year all-cause mortality, sepsis progression, and atherosclerosis incident risk before clinical symptoms manifest. Conversely, the MetaboHealth score is recognized as vastly superior for forecasting long-term biological aging and evaluating comprehensive, organismal physiological decline over decades.

Strengths and Weaknesses

Metabolic Vulnerability Index (MVX)

  • Strengths: MVX is commercially accessible directly to consumers in the United States (e.g., via Labcorp for approximately $35 to $40). It provides a highly actionable 0–100 score that isolates chronic vulnerability from acute inflammatory noise. Baseline MVX scores are independent of chronological age, making it an effective tool for identifying early-stage metabolic strain in ostensibly healthy younger adults.

  • Weaknesses: The “malnutrition” nomenclature used for the MMX sub-score is clinically counterintuitive for modern populations, where the primary phenotype is cellular undernourishment driven by caloric overconsumption. Furthermore, the absence of explicit glycemic markers (like glucose or HbA1c) requires patients to order supplementary metabolic panels to achieve a complete clinical picture.

Personal-MetaboHealth

  • Strengths: This panel offers a superior framework for tracking longevity interventions and lifestyle modifications. Because it captures glycemic control, liver function, and dietary lipid quality, it has been successfully validated to track measurable physiological reversals following targeted 3-month metabolic interventions.

  • Weaknesses: It lacks a direct-to-consumer commercial pathway in the United States. Additionally, the exact statistical weights applied to the 10 biomarkers are proprietary, restricting independent clinical reproduction and peer review of the raw algorithm.

Knowledge Gaps and Required Data

For both scoring systems, the precise causal pathways driving the algorithm outputs remain largely speculative. While they excel at stratifying risk, it is unknown if these biomarkers are drivers of aging or downstream consequences. Extensive randomized controlled trials are required to determine whether targeted pharmacological interventions—such as suppressing BCAAs with mTOR inhibitors (e.g., rapamycin) or lowering GlycA with senolytics—result in proportional lifespan extension, or if these interventions merely alter the biomarker score without reversing fundamental biological aging.