The study, published in European Geriatric Medicine by Junghyeon Hyeon of the Department of Family Medicine at Busan Paik Hospital, Inje University College of Medicine, examined data from the 2024 Korea National Health and Nutrition Examination Survey (KNHANES). Among 1,178 adults aged 65 and older who underwent both DXA and BIA scans on the same survey cycle—the survey-weighted mean age was 72.1 years and 54.3 percent were women—the agreement between the two devices proved far weaker than many clinicians might assume. Under the combined height-or-body-mass-index criterion adopted by the Asian Working Group for Sarcopenia (AWGS) in its 2025 consensus update, 83.0 percent of participants were classified as having low muscle mass by DXA, compared with just 55.5 percent by BIA. Observed agreement between the two modalities was only 65.4 percent, and Cohen’s kappa—a statistic that corrects for chance agreement—was a mere 0.246, a level conventionally interpreted as fair at best.

The technical heart of the problem lies in how each device estimates skeletal muscle and how the resulting values are adjusted for body size. DXA directly quantifies lean soft tissue by distinguishing X-ray attenuation at two energy levels, providing a regional partition of appendicular lean mass in the arms and legs. BIA, by contrast, does not measure muscle at all; it infers total body water from the resistance of tissues to a weak alternating electrical current and then applies population-specific prediction equations to estimate appendicular lean mass. Because fat and muscle conduct electricity differently, and because hydration status, body geometry and the equations themselves all influence the estimate, BIA values can diverge systematically from DXA references. When those estimates are then normalized—either by height squared, as in the traditional appendicular lean mass index, or by body mass index (BMI), a newer adjustment intended to account for adiposity—the divergence cascades into large differences in who crosses the diagnostic threshold.

The choice of adjustment criterion turned out to be decisive. Agreement was best under the BMI-adjusted criterion, where 50.6 percent of participants met the low-muscle-mass threshold by DXA and 41.1 percent by BIA, yielding 79.8 percent observed agreement and a kappa of 0.598—moderate, though still short of interchangeability. Agreement was worst under the height-adjusted criterion: 70.8 percent of older Koreans were flagged as having low muscle mass by DXA versus only 28.8 percent by BIA, with 56.5 percent observed agreement and a kappa of 0.258. The combined height-or-BMI definition, which classifies a person as having low muscle mass if either threshold is crossed, did not reconcile the two devices; instead, it widened the gap, because DXA’s height-adjusted flag swept large numbers of people into the low-muscle-mass category that BIA never reached.

The implications sharpen considerably when muscle strength enters the equation, since sarcopenia under AWGS 2025 requires not only low muscle mass but also low physical performance or low muscle strength—most commonly assessed by handgrip dynamometry. Among the 1,121 participants with handgrip-strength data, 131 had low handgrip strength. Within this clinically consequential subgroup, combined-criterion low muscle mass was found in 93.1 percent by DXA and 80.3 percent by BIA, but the devices disagreed in 22 individuals: 18 were classified as having low muscle mass by DXA but not BIA, and 4 by BIA but not DXA. In other words, for the very patients in whom a sarcopenia diagnosis would actually be made, the choice of device could change the answer.

The headline sarcopenia figures, by contrast, looked deceptively harmonious. Prevalence was 10.0 percent by DXA and 8.7 percent by BIA, with a kappa of 0.895—almost perfect agreement. But the study reveals why that harmony is partly illusory: 990 of the participants had normal handgrip strength and were therefore concordantly classified as non-sarcopenic regardless of what their muscle scans showed. The apparent consensus between the devices was driven overwhelmingly by the shared strength gate rather than by genuine agreement about muscle mass. Strip away the strength requirement, and the discordance re-emerges in full force among those who fail it.

These findings arrive at a pivotal moment. The AWGS 2025 consensus update, published in Nature Aging, marked a deliberate shift in framing—from sarcopenia as a discrete disease toward a broader concept of muscle health—and introduced more flexible body-size adjustments, including the BMI-based option, in recognition of the peculiarities of Asian populations, among whom high adiposity at lower BMI values can mask or distort muscle assessments. The 2019 AWGS criteria, like their European counterparts, had anchored low muscle mass primarily to the height-adjusted appendicular lean mass index, a metric that the new analysis suggests is especially vulnerable to device-dependent classification: it produced the largest DXA–BIA gap of any definition tested.

Prior research has already hinted at this fragility. The Bunkyo Health Study in Japan, which compared BIA and DXA for appendicular lean mass in more than 1,600 community-dwelling older adults, documented systematic differences between the modalities, and a 2024 systematic review and meta-analysis of muscle-measurement techniques concluded that agreement varies substantially depending on the device, the derived variable, and the cutoffs applied. Work from hospital cohorts has similarly found that multifrequency BIA can overestimate or underestimate muscle mass relative to DXA depending on patient characteristics. What the new Korean analysis adds is a nationally representative test of the newest AWGS 2025 criteria themselves—including the combined height-or-BMI definition—showing that even the updated framework does not erase modality dependence.

For clinicians and researchers, the practical message is one of precision and transparency. A statement that a patient “has low muscle mass” is incomplete, and potentially misleading, unless it specifies the measurement device, the derived muscle variable—height-adjusted or BMI-adjusted appendicular lean mass—and the adjustment criterion used to apply the cutoff. In screening and epidemiological surveys, where portable BIA dominates because it is inexpensive, radiation-free and quick, prevalence estimates of low muscle mass may be substantially lower than what DXA-based studies of the same population would report. In clinical practice, a patient flagged by DXA but cleared by BIA may miss an opportunity for early resistance training and nutritional intervention, while the reverse discordance could prompt unnecessary concern. The study’s author argues that interpretation should always be integrated with muscle-strength assessment, since strength data—handgrip in this case—anchor the diagnosis in functional terms that matter to patients.

The analysis is not without limits inherent to its design. Cross-sectional survey data capture agreement at a single point in time and cannot address whether one device better predicts falls, fractures, disability or mortality—the outcomes that give sarcopenia its clinical meaning. The KNHANES 2024 protocol was approved by the Institutional Review Board of the Korea Disease Control and Prevention Agency, and the study used de-identified public data with survey weights applied to prevalence estimates while agreement statistics were computed unweighted. Still, with populations across Asia and beyond aging rapidly, and with sarcopenia prevalence projected to climb, the findings sound a timely warning: before the field can agree on how many older adults have sarcopenia, it must first agree on how to measure the muscle they have lost. Until then, DXA and BIA should be treated as complementary tools with distinct biases—not as interchangeable windows onto the aging body.

Cancer survivorship is a complex journey impacted not only by the type and stage of cancer but also by lifestyle habits and underlying genetic predispositions. In particular, obesity has emerged as a significant modifiable risk factor that may increase the chances of cancer recurrence and adversely affect survival rates. This study, therefore, focused on evaluating how various demographic, health, and nutritional factors interact with BMI in individuals who have previously been diagnosed with cancer, but who are not in the terminal stage of their illness.

Utilizing data from the 7th and 8th Korea National Health and Nutrition Examination Survey (KNHANES), spanning 2016 to 2020, the research team undertook a secondary analysis encompassing over four million cancer survivors. This vast dataset enabled the investigators to perform robust statistical analyses and identify patterns that might otherwise remain obscured in smaller studies. The inclusion criteria ensured participants were either in remission after initial cancer management or undergoing treatment for advanced stages, excluding those in terminal conditions, thus providing a representative cohort of ongoing survivorship.

The methodology entailed applying complex sampling techniques combined with both descriptive and inferential statistics. Tools such as cross-tabulation, chi-square tests, t-tests, as well as multivariable linear regression models were employed to parse out the significant determinants influencing BMI in this unique population. The utilization of IBM SPSS software facilitated rigorous data management and analysis, ensuring the credibility and reproducibility of the findings.

Results from this analysis illuminated several demographic factors closely linked to BMI variations among cancer survivors. Notably, gender emerged as a significant determinant, with the data indicating differential BMI distributions between males and females. Marital status and engagement in economic activities also played pivotal roles, highlighting the intricate social dimensions influencing body weight management post-cancer diagnosis. These variables underscore the importance of considering personal and social contexts when designing supportive care interventions.

From a health perspective, the presence of hypertension was significantly associated with higher BMI values, which aligns with existing literature on the interplay between obesity and cardiovascular risk factors in cancer populations. Additionally, hemoglobin levels, an indicator often reflective of overall physiological well-being, exhibited a strong correlation with BMI, suggesting that metabolic and nutritional statuses are interconnected factors influencing cancer survivors’ health profiles.

One of the more novel aspects of the study pertains to nutrition-related characteristics. Frequency of breakfast consumption was found to have a statistically significant inverse relationship with BMI, emphasizing the potential impact of consistent meal routines on maintaining healthy body weight. Moreover, micronutrient intake, specifically vitamin D and vitamin C, was differentially associated with BMI levels, implying that nutrient adequacy and dietary composition are critical components in survivorship care plans.

Interestingly, sodium intake did not show a statistically meaningful relationship with BMI in this cohort, a finding that may provoke further inquiry into the nuanced dietary influences on weight regulation in cancer patients. These nutrition-centric insights advocate for personalized dietary guidelines tailored to cancer survivors’ unique metabolic demands and risk profiles.

The broader implications of these results testify to the critical need for integrated health strategies aimed at assessing and managing BMI among cancer survivors. Since obesity can potentiate cancer recurrence and complicate treatment responses, healthcare providers should prioritize establishing comprehensive behavioral and nutritional interventions. Such measures not only promote weight optimization but are likely to improve overall quality of life and extend healthy survival.

The study’s authors stress that BMI management is not a mere standalone goal but must be embedded within multi-dimensional approaches that consider psychological, social, and clinical facets of survivorship. Encouraging routine health assessments, tailored counseling, and continuous monitoring can empower survivors to adopt sustainable health habits, thereby mitigating future metabolic or oncological complications.

To translate research insights into practice, development of targeted health intervention programs becomes imperative. These should be evidence-based, culturally sensitive, and adaptable to diverse survivor populations to maximize adherence and effectiveness. Moreover, policy frameworks must support accessible health resources that integrate nutritional education and physical activity promotion within survivorship care pathways.

While this study advances understanding, it also opens avenues for future research to explore causal mechanisms underpinning the observed associations. Particularly, longitudinal studies tracking changes in BMI and concomitant health behaviors over time could elucidate critical windows for intervention. Further exploration into genetic markers interacting with lifestyle factors may also refine personalized survivorship strategies.

In conclusion, the robust analysis conducted on a large national dataset confirms the complex interrelations between BMI and a spectrum of demographic, health, and nutritional factors in cancer survivors. These data reinforce the urgency for comprehensive survivorship care models that proactively address obesity to enhance long-term outcomes. Health professionals, researchers, and policymakers alike should harness these findings to foster innovations that support cancer survivors in achieving sustained wellness.

Given the projected increase in global cancer survivorship, the integration of BMI management into standard oncological follow-up represents a timely and vital endeavor. By championing multidisciplinary approaches that encapsulate medical, behavioral, and societal determinants of health, the community can collectively improve survival quality for this growing population.

As the scientific community continues to elucidate the nuanced interactions between obesity and cancer outcomes, this study exemplifies the power of leveraging large-scale epidemiological data to inform clinical and public health strategies. Its findings resonate as a clarion call to reimagine survivorship care in the context of modern chronic disease management.

The future of cancer survivorship hinges on our ability to adopt personalized, data-driven interventions that recognize the heterogeneity of survivors’ experiences. Effective BMI control, guided by comprehensive risk assessments and tailored support, will serve not only to reduce recurrence risk but also to fortify overall health resilience.

Ultimately, bridging research with actionable health policies will catalyze progress toward elevating the standard of care for cancer survivors globally. This landmark analysis published in BMC Cancer marks a fundamental step forward in that mission, illuminating pathways toward healthier survivorship for millions worldwide.

Subject of Research: Health-related characteristics and factors associated with body mass index in cancer survivors.

Article Title: Health-related characteristics of Cancer survivors according to body mass index: a secondary data analysis.

Article References:
Ku, I.H., Ko, S. Health-related characteristics of Cancer survivors according to body mass index: a secondary data analysis.
BMC Cancer 25, 865 (2025). https://doi.org/10.1186/s12885-025-13871-0

Image Credits: Scienmag.com

DOI: https://doi.org/10.1186/s12885-025-13871-0