In older adults with type 2 diabetes (T2D), low BMI was associated with a higher risk for death in a new study — a finding that challenges the conventional wisdom that lower weight leads to superior health outcomes.

The association appeared to be mediated by lower lipid levels, particularly high-density lipoprotein (HDL) cholesterol.

“Low BMI in this population rarely reflects simple underweight status but more often coexists with sarcopenia, abnormal fat distribution, and depleted energy reserves,” Hongfei Mo, of the School of Public Health and the Fudan Development Institute at Fudan University, Shanghai, China, and colleagues wrote in their study, published online in Diabetology & Metabolic Syndrome.

These are all signs of “altered body composition,” which are “associated with accelerated aging- and diabetes-related pathophysiological processes and ultimately with elevated mortality risk,” the authors explained.

The findings “highlight the urgent need” to address low BMI in the management of older adults with T2D “as current guidelines tend to overemphasize obesity treatment and overlook the risks associated with underweight status,” they added.

Hallmark Physiologic Change

Involuntary weight loss and low BMI in older adults are “key factors” influencing the prognosis of T2D, the authors noted. In fact, recent research suggests that maintaining a moderate body weight is significantly associated with better survival outcomes in this population.

Dyslipidemia, a “hallmark physiological change in diabetes,” may be implicated in the association between low BMI and mortality risk because “core lipid parameters” (total cholesterol, triglycerides, low-density lipoprotein [LDL] cholesterol, and HDL cholesterol) are “directly involved in energy homeostasis, inflammatory, and cell membrane stability,” the researchers explained.

Therefore, they set out to shine a light on the role of dyslipidemia in mediating the relationship between low BMI (< 22 ) and all-cause mortality in older adults (aged ≥ 60 years).

The current study included data from 372,559 participants (45.01% male, mean age 72.04 ± 7.31 years) from the Shanghai Municipal Center for Disease Control and Prevention, who were observed between May 1, 2020, and December 31, 2024. Of these, 76,326 were classified as having low BMI. The median follow-up was 19 months.

Age, sex, education, smoking, alcohol, and physical activity were included as confounding variables in regression and mediation models because these differed between the groups (P < .001 for all).

Lipid Parameters and Low BMI

Linear regression analysis found that low BMI was associated with all four lipid parameters. After adjustment for confounding variables, the associations were total cholesterol, beta = 0.08 (95% CI, 0.07-0.09); triglycerides, beta = -0.17 (95% CI, -0.19 to -0.15); HDL cholesterol, beta = 0.10 (95% CI, 0.10-0.11); LDL cholesterol, beta = 0.03 (95% CI, 0.01-0.04); P < .001 for all.

After adjustment, low BMI was found to be associated with a statistically significant risk for all-cause mortality (hazard ratio, 1.10; 95% CI, 1.06-1.13) compared with individuals with higher BMI.

Lipid parameters also showed nonlinear relationships with mortality risk. Total cholesterol and LDL cholesterol followed U-shaped associations, whereas HDL cholesterol showed an L-shaped pattern. The optimal thresholds that emerged in restricted cubic spline models were 5.03 mmol/L for total cholesterol, 0.98 mmol/L for HDL cholesterol, and 3.39 mmol/L for LDL cholesterol.

“This indicates that both excessively high and low concentrations of [total cholesterol] and LDL [cholesterol] are associated with mortality,” the authors commented. “For HDL [cholesterol], a floor effect is observed: Below a certain threshold, risk escalates sharply,” they continued.

Notably, the optimal HDL cholesterol threshold identified in this cohort (0.98 mmol/L) is lower than commonly accepted normal lower limits (typically > 1.0 mmol/L for men and > 1.3 mmol/L for women), they explained.

“This suggests that even slight reductions in HDL [cholesterol], within the lower end of the normal range, may signal increased mortality for T2D older adults with low BMI.”

In mediation analyses, total cholesterol, HDL cholesterol, and LDL cholesterol demonstrated significant negative mediation effects on the association between low BMI and mortality. HDL cholesterol showed the most pronounced masking effect, accounting for a proportion mediated of -111.99% and a beta-average causal mediation effect of 1.44 (95% CI, 1.33-1.56). Total cholesterol and LDL cholesterol showed smaller negative mediation effects (proportions mediated, -16.68% and -4.12%, respectively).

Limitations include the potential for residual confounding despite multivariable adjustment and the absence of data about body composition and direct measures of muscle mass, fat mass, and their distribution, the study authors noted.

Moreover, “from a public health perspective, diabetes management programs should increase focus on low BMI and underweight older adults by strengthening screening, monitoring, and developing targeted interventions to reduce disease burden in this vulnerable group,” they wrote.

They recommended optimizing comorbidity management “to reverse metabolic wasting.”

Low BMI Doesn’t Equate to Low Adiposity

Commenting for Medscape Medical News, Scott Isaacs, MD, adjunct assistant professor of medicine, Emory University School of Medicine, Atlanta, said that in the setting of this population, “being very thin may mean the body is losing muscle as part of overall poor health and frailty, and cholesterol levels are becoming abnormal.”

“Taken together, these findings suggest that in older people with T2D, the priority is maintaining a healthy weight and preserving lean muscle mass to prevent sarcopenia and deconditioning,” said Isaacs, immediate past president of the American Association of Clinical Endocrinology, who wasn’t involved with the study.

Also commenting for Medscape Medical News, Amy Rothberg, MD, clinical professor of medicine, Division of Metabolism, Endocrinology and Diabetes, Department of Internal Medicine, University of Michigan, Ann Arbor, Michigan, noted that the thresholds for diagnosing overweight and obesity in the Asian population differ from those of non-Asians.

“Asian people tend to have greater adiposity for every level of BMI, but BMI thresholds are based on a Caucasian reference,” she said, so these findings may not be applicable to other populations.

In older populations, a BMI of 22 may not have the same implications as in younger people, added Rothberg, research professor of nutritional sciences in the School of Public Health, who also wasn’t involved with the study.

“You need to measure waist circumference because the person may have lost lean muscle mass but still have excess adipose tissue, including central adiposity — a surrogate for visceral adiposity that’s likely in those with diabetes,” she explained.

She cautioned, however, that while physicians should be careful about excessive weight loss in older adults, they shouldn’t exclude weight loss as one of several interventions to reduce associated comorbidities.

The study was funded by the Noncommunicable Chronic Diseases-National Science and Technology Major Project. The authors and Isaacs reported having no relevant financial relationships. Rothberg reported engaging in speaking engagements for Eli Lilly and Company on the subject of diabetes remission.

Batya Swift Yasgur, MA, LSW, is a freelance writer with a counseling practice in Teaneck, New Jersey. She is a regular contributor to numerous medical publications, including Medscape and WebMD, and is the author of several consumer-oriented health books as well as Behind the Burqa: Our Lives in Afghanistan and How We Escaped to Freedom (the memoir of two brave Afghan sisters who told her their story).