In this nationally representative analysis of older Indian adults, we compared the obesity burden identified using conventional BMI criteria with that classified using the revised pre-clinical and clinical obesity framework and present some interesting findings. While the prevalence of high BMI and pre-clinical obesity was broadly similar, the prevalence of clinical obesity was substantially lower, suggesting that the revised framework may help distinguish excess adiposity from obesity accompanied by multimorbidity and/or functional limitation. We also observed marked geographic and sociodemographic disparities in both obesity states. Women, urban residents, and individuals with higher education and wealth had consistently higher odds of living with both pre-clinical and CO. In addition, participants with pre-clinical obesity more often reported excellent self-rated health, whereas poor self-rated health was more common among those with CO, highlighting a potential mismatch between perceived and clinically relevant obesity burden.

A substantial proportion of participants also showed elevated anthropometric indicators of adiposity, including WC, WHR, and WHtR. This pattern is consistent with previous evidence from Asian populations showing that central adiposity and metabolic risk may occur at relatively lower BMI levels [18]. Such findings support the continued use of ethnicity-sensitive anthropometric thresholds in Indian populations. Ethnic variations in BF% distribution and metabolic characteristics can be attributed to gene variants (e.g., FTO, ADIPOQ, TCF7L2) and epigenetic modifications that predispose the Asian population to greater visceral fat accumulation, lower insulin sensitivity, and diminished beta-cell function, exacerbated by lower adiponectin and higher cortisol levels [19,20,21,22]. A lower capacity for peripheral fat storage leads to earlier deposition of excess fat in the visceral region, which is more metabolically active than subcutaneous fat and is associated with pro-inflammatory cytokines (e.g., TNF-α, IL-6) and free fatty acid (FFA) release [23, 24]. The lower muscle mass reduces glucose uptake capacity (via GLUT4 receptors), promoting hyperglycemia and fat storage. There is a higher tendency to develop ectopic fat more readily, contributing to the early onset of non-alcoholic fatty liver disease (NAFLD), metabolic derangements, and enhancing cardiovascular risk even at lower obesity levels [25,26,27]. Such phenotypic predispositions underscore the need for ethnicity-specific cut-offs for obesity indicators and targeted interventions to address central obesity and its associated health risks in different populations [28]. At the same time, our results suggest that anthropometric burden alone does not fully capture clinically relevant obesity, as only a subset of participants identified under the revised framework met criteria for CO. This distinction may be important for population triage and risk stratification, especially in settings where obesity-related morbidity and functional impairment need to be prioritized.

In our study, clinical obesity was operationalized as multimorbidity and ADL limitation among participants who met the anthropometric criteria. This approach reflects the intent of the revised framework to identify obesity accompanied by clinical compromise, although it remains constrained by the variables available in LASI. We observed that a higher proportion of our participants had multiple co-existing morbidities. Globally, multimorbidity is a growing concern. A global systematic review reported a pooled prevalence of 37.2% (95% CI = 34.9–39.4%) for multimorbidity, ranging from 45.7 to 35% across different geographical locations [29]. The study showed a higher prevalence among women (39.4%) than men (32.8%). More than half of the adult population worldwide aged ≥ 60 had multimorbid conditions (51.0%). Another multi-country analysis of data from the Collaborative Research on Ageing in Europe project (Finland, Poland, and Spain) and the World Health Organization’s Study on Global Aging and Adult Health (China, Ghana, India, Mexico, Russia, and South Africa) reported the highest prevalence of multimorbidity in Russia (71.9%), whereas China (45.1%) and Ghana (48.3%) had the lowest [30]. In India, the prevalence of multimorbidity among older adults varies between 28.3 and 57.9% because of methodological differences, with striking variations across sociodemographic characteristics, including increasing age, women, urban residence, and living alone [30, 31]. Likewise, in India, studies have reported varying prevalence rates of ADL disabilities among the elderly, ranging between 3 and 6%. Globally, the prevalence of ADL impairments varies across regions. In the US, the prevalence of severe to extreme ADL impairment increased from 2.8% in the 40–44 age group to 27.3% in the oldest age group. ADL worsening is significantly linked to women, and older individuals are more likely to experience ADL impairments. Factors such as lack of physical activity were significantly associated with higher risks of severe ADL and IADL disabilities. Our findings concerning multimorbidity and limitations of ADL should therefore be interpreted as a pragmatic population-level approximation of clinical obesity rather than a complete capture of all obesity-related organ dysfunction.

The marked subnational variation observed in our study underscores the need for region-specific approaches to obesity surveillance and intervention. The disparities have been attributed to several factors, including the urban-rural divide, and such disparities have widened over time [17]. Urban lifestyles often involve reduced physical activity and increased consumption of calorie-dense foods, contributing to higher obesity rates [32, 33]. However, rural areas are increasingly exposed to unhealthy lifestyles because of more mechanization of farming and increased frequency of junk food consumption, which is a cause of concern [34]. Cultural dietary habits significantly influence obesity prevalence. For instance, southern states like Kerala and Tamil Nadu report higher abdominal obesity rates, potentially due to regional dietary patterns rich in fats and sugars [35]. Regions with better healthcare access and Health literacy may have higher reported obesity rates due to increased diagnosis and reporting. Conversely, areas with limited healthcare access might underreport obesity prevalence.

An important finding was the contrast between self-rated health across the two obesity states. Participants with pre-clinical obesity more often reported excellent health, whereas poorer self-rated health was more common among those with CO. This pattern suggests that excess adiposity may remain underrecognized when not yet accompanied by overt morbidity or functional decline [36]. Eventually, co-existing morbidities trigger concerns about social rejection and the internalization of weight stigma, which increases psychological distress and decreases the self-rated scores [37]. Our results are in concordance with some previous studies showing that increasing obesity aggravates limitations in ADLs, multimorbidities, and chronic pain, contributing to worsened self-rated health [6]. Such findings necessitate a wider advocacy to de-normalize obesity as a status symbol for prosperity and increase awareness about the health risk, despite not having any overt symptoms or functional limitations [38]. Our study also reiterates the lack of physical activity as a significant risk factor for both obesity states. While we have sufficient evidence to demonstrate the health benefits of physical activity, we cannot assess the reasons for insufficient physical activity levels because of data constraints. Still, it is essential to raise concerns about an enabling environment, which is integral to encouraging physical activity. Despite government-endorsed guidelines for different age groups, adoption remains a challenge due to insufficient attention to the built environment [39].

This study has several strengths and limitations. Its main strengths lie in applying the revised obesity framework to a large, nationally representative sample of older Indian adults and in presenting comparative estimates alongside conventional BMI-based classification. The dataset also allowed examination of subnational variation and key sociodemographic correlates. However, the findings must be interpreted in light of important limitations. First, the analysis was based on LASI Wave 1 (2017–2018) and may not reflect the current burden of obesity. Second, although anthropometric measurements were directly recorded, chronic conditions and ADL limitations were self-reported and may be affected by recall error, reporting bias, and differential access to diagnosis. Third, the operational definition of CO was constrained by dataset availability; important obesity-related conditions such as sleep apnea, heart failure, renal insufficiency, and steatotic liver disease were not available. Fourth, proxy anthropometric measures may misclassify body fat distribution and metabolic risk. Also, chronic conditions and ADL limitations were self-reported and may have been affected by recall error, reporting bias, underdiagnosis, and limited variable availability in LASI. Fifth, unhealthy diet could not be assessed because LASI did not include the required information. Finally, given the cross-sectional design, temporal relationships cannot be inferred. Lastly, some regional and state/UT-specific estimates had wide confidence intervals, likely reflecting smaller sample sizes in those strata; these findings should therefore be interpreted with caution.

Despite these limitations, the findings have important implications. The revised framework may offer a more clinically stratified view of obesity burden than BMI alone by separating excess adiposity without overt compromise from obesity accompanied by morbidity and functional limitation. This may be useful for population triage, health communication, and service prioritization, particularly in resource-constrained settings. At the population level, the revised framework may offer a pragmatic way to estimate clinically relevant obesity burden using variables feasible for large surveys. However, it cannot capture the full spectrum of obesity-related organ dysfunction identifiable in specialist clinical settings. This distinction is important, as obesity estimates derived from community surveys and clinic-based evaluations are likely to differ in both scope and interpretation. Importantly, such classification may support stepwise prioritization of anti-obesity pharmacotherapy by directing limited treatment capacity toward those with the highest clinical burden and the greatest likelihood of benefit, rather than relying solely on BMI. The observed urban, affluent, and regional gradients also support the need for geographically and socially tailored prevention strategies. In addition, the mismatch between self-rated health and revised obesity status suggests that public awareness efforts should address obesity-related risk even before clinical compromise becomes evident. Our findings also indicate that estimates of clinical obesity are likely shaped not only by underlying disease burden but also by healthcare access, diagnosis, and functional assessment. Populations with better access to screening and chronic disease care may be more likely to be classified as having clinical obesity because multimorbidity is more readily detected, whereas limited access may underestimate the true burden. Similarly, access to rehabilitation access may influence the extent to which functional limitations are identified and managed. These considerations reinforce the need to strengthen primary care, chronic disease detection, and rehabilitation services when interpreting and applying revised obesity classifications in public health settings.