Routine Blood Test Marker RDW Linked to Higher Hip Fracture Risk in Older Adults

A routine blood test that costs pennies and is performed millions of times a day in clinics around the world may carry a hidden warning about broken bones. New research drawing on the UK Biobank suggests that red blood cell distribution width, a standard measure of the variation in the size of circulating red blood cells, is associated with the risk of hip fracture in older adults. The finding, published in BMC Geriatrics, adds a hematologic twist to the search for better ways to identify who is most likely to suffer one of the most devastating injuries of later life.

The study, led by Kaihua Qin and Shuhuai Tan of the Medical School of Chinese PLA and the Fourth Medical Center of the Chinese PLA General Hospital in Beijing, together with colleagues including corresponding authors Wen Ju and Houchen Lyu, took advantage of one of the largest biomedical databases in existence. The team identified 76,078 UK Biobank participants aged 65 to 73 years who had no prior history of hip fracture at recruitment. For each participant, they extracted the baseline red blood cell distribution width from the complete blood count measured when the volunteer joined the study, and then tracked health records over a median follow-up of 14.06 years.

Red blood cell distribution width, usually abbreviated RDW, is reported as the coefficient of variation of red cell volumes. In plain terms, it quantifies how much the individual cells in a blood sample differ in size. Clinicians have long used it to help distinguish types of anemia, since conditions such as iron deficiency produce a mixed population of large and small cells and therefore a wider distribution. What has attracted growing scientific interest in recent years, however, is the observation that a higher RDW tends to track with aging, chronic inflammation, nutritional deficits, and a range of adverse outcomes far beyond the bloodstream, including cardiovascular disease and mortality.

During the roughly fourteen years of observation, 2,016 of the participants suffered an incident hip fracture, corresponding to an incidence rate of 1.99 per 1,000 person-years, with a 95 percent confidence interval of 1.90 to 2.08. That baseline rate is itself a reminder of the scale of the problem: hip fractures are among the most consequential injuries in geriatric medicine, frequently triggering a cascade of immobility, loss of independence, and excess mortality. Any marker that could flag elevated risk before the fall occurs would be clinically valuable, particularly if it came free with a test that is already being ordered for other reasons.

To isolate the contribution of RDW, the researchers used multivariable Cox proportional hazards models, the standard statistical machinery for time-to-event cohort data. The models adjusted for a panel of relevant confounders, and missing covariate data were handled with multiple imputation, a technique that fills in gaps in the dataset in a way that properly propagates the uncertainty introduced by the missing values. The result was strikingly consistent: each 1-percentage-point increase in RDW was associated with an 8 percent higher hazard of hip fracture, with a hazard ratio of 1.08 and a 95 percent confidence interval of 1.03 to 1.13, a relationship that reached statistical significance at P less than 0.001.

A crucial technical question in any dose-response analysis of this kind is whether the association is linear or whether risk climbs only above some threshold. The team addressed this with restricted cubic splines, a flexible modeling approach that fits smooth curves through the data without forcing a straight line. The spline analysis supported an overall association between RDW and hip fracture risk, with P-overall equal to 0.001, but found no evidence of nonlinearity, with P-nonlinearity at 0.448. In other words, the relationship appears to behave like a steady gradient: the wider the spread of red cell sizes, the greater the fracture hazard, without any obvious safe zone or tipping point within the observed range.

The investigators also probed whether the association differed across clinically meaningful subgroups. Because anemia itself could plausibly confound or modify the relationship, and because fracture epidemiology differs substantially between women and men, they stratified the analysis by sex and by anemia status and formally tested for interaction. Neither test revealed effect modification: the P value for interaction was 0.795 for sex and 0.532 for anemia status. The association between higher RDW and higher hip fracture risk therefore held in both sexes and independently of whether participants were anemic, suggesting that RDW captures something beyond the presence of a low hemoglobin level.

What could connect the size variability of red blood cells to the structural integrity of the hip? The authors and the broader literature point to several plausible mechanisms. An elevated RDW is a nonspecific but sensitive indicator of physiological stress: it rises with chronic inflammation, oxidative stress, impaired iron metabolism, and nutritional deficiencies, including deficits in iron, vitamin B12, and folate. Each of these states has been implicated in bone health, whether through direct effects on osteoblast activity, through inflammatory cytokines that accelerate bone resorption, or through frailty pathways that also increase fall risk. In this sense, RDW may function as an integrative barometer of biological aging, summarizing in a single number the cumulative wear on systems that also govern the skeleton.

The study has the strengths and limitations inherent to its design. Its prospective structure, large sample, long follow-up, and use of a nationally representative research cohort give it considerable power, and the adjustment for measured confounders with modern missing-data methods strengthens causal inference. Yet RDW is a marker, not a proven cause, and residual confounding by unmeasured factors such as detailed dietary intake, physical activity, or medication use cannot be excluded. The cohort was restricted to UK Biobank participants aged 65 to 73, a relatively healthy and predominantly European-ancestry group, so generalization to frailer or more diverse populations will require confirmation. The authors are careful to frame RDW as a candidate component of fracture risk prediction models that warrants further evaluation, not as a standalone diagnostic.

Even with those caveats, the practical appeal of the finding is hard to overstate. Fracture risk assessment currently relies heavily on bone mineral density and clinical risk factors, tools that miss a substantial share of people who go on to fracture. A hematologic index that is already measured at essentially every routine checkup, that behaves linearly, and that applies to both sexes regardless of anemia status could be folded into existing prediction algorithms at essentially no additional cost or burden. The work, funded by the National Key Research and Development Program of China, the National Natural Science Foundation of China, and the Capital’s Funds for Health Improvement and Research Program, was conducted under UK Biobank Application Number 105398. If future studies validate the signal in independent cohorts and demonstrate that incorporating RDW genuinely improves risk classification, one of medicine’s most ordinary numbers may earn an unexpected second life in the fight against hip fractures.

Subject of Research: Association between red blood cell distribution width and hip fracture risk in older adults

Article Title: Red blood cell distribution width and risk of hip fracture in older adults: a UK Biobank prospective cohort study

Article References: Qin, K., Tan, S., Jiang, Y., Luo, Y., Liu, X., Yang, C., Liang, D., Zhang, L., Cui, X., Ju, W., & Lyu, H. (2026). Red blood cell distribution width and risk of hip fracture in older adults: a UK Biobank prospective cohort study. BMC Geriatrics. https://doi.org/10.1186/s12877-026-08400-y

Image Credits: AI Generated

DOI: 10.1186/s12877-026-08400-y

Keywords: red blood cell distribution width, hip fracture, UK Biobank, older adults, cohort study, anemia, fracture risk prediction, hematology, osteoporosis, geriatrics, Cox proportional hazards, epidemiology