The research, published in BMC Endocrine Disorders, was led by Xuanrong Chen and Hongrong Deng of the Third Affiliated Hospital of Sun Yat-Sen University in Guangzhou, together with colleagues from Sun Yat-Sen University and Shenzhen Bao’an District People’s Hospital. The team set out to answer a deceptively simple question: among adults with diabetes in Guangdong Province, do the dietary patterns people actually eat—rather than individual nutrients in isolation—predict how well they achieve their metabolic goals? The answer, drawn from a large cross-sectional analysis, is a qualified yes, with the strongest and most consistent signal appearing for blood pressure control.

To conduct the study, the researchers turned to data from the China National Diabetic Chronic Complications Study, a nationwide effort registered in the Chinese Clinical Trial Registry in January 2018 under identifier ChiCTR1800014432. From this parent study, the team performed a secondary cross-sectional analysis of participants with diabetes mellitus recruited from four regions of Guangdong Province, a densely populated economic powerhouse in southern China where dietary traditions differ markedly from those of the country’s north. After applying eligibility criteria, 1,465 participants remained in the final analytical sample, of whom 48.19 percent were male—a near-even gender split that strengthens the generalizability of the findings within this population.

Dietary intake was assessed using a semi-quantitative food frequency questionnaire, a standard epidemiological tool that asks participants how often they consume specific foods and in what typical portions. Rather than examining single foods or nutrients one at a time—an approach that can miss the synergistic effects of whole diets—the researchers employed principal component analysis, or PCA, a statistical technique that identifies underlying patterns of food consumption from the correlation structure of the dietary data. PCA works by condensing dozens of correlated food variables into a small number of independent components, each representing a cluster of foods that tend to be eaten together. In this study, the analysis surfaced a so-called prudent dietary pattern, a term used in nutrition science to describe eating habits rich in vegetables, fruits, whole grains, and other foods conventionally associated with health benefits, in contrast to a Western pattern dominated by refined grains, red and processed meats, and sugar-sweetened beverages.

With the dietary pattern quantified, each participant received a score reflecting their degree of adherence, and the sample was divided into tertiles—three groups of roughly equal size representing low, intermediate, and high adherence. The researchers then examined whether adherence level predicted attainment of three internationally recognized metabolic targets: glycated hemoglobin below the threshold recommended for diabetes control, blood pressure within guideline limits, and low-density lipoprotein cholesterol at goal. Multivariable logistic regression was used to estimate the odds of achieving each target across adherence tertiles, with statistical adjustment for potential confounders such as age, sex, body mass index, disease duration, medication use, and other lifestyle factors.

The results revealed a nuanced picture. Compared with participants in the lowest tertile of prudent dietary pattern adherence, those in the intermediate tertile had significantly higher odds of reaching their glycated hemoglobin target, with an adjusted odds ratio of 1.428 and a 95 percent confidence interval of 1.074 to 1.898, corresponding to a p-value of 0.014. Meanwhile, participants in the highest tertile of adherence showed significantly higher odds of achieving their blood pressure target, with an adjusted odds ratio of 1.439 and a 95 percent confidence interval of 1.042 to 1.989, and a p-value of 0.027. In plain terms, people who ate in line with the prudent pattern were roughly 40 percent more likely to hit these individual targets than those at the bottom of the adherence distribution, after accounting for other factors that influence metabolic control.

Not every outcome followed the same script. The researchers found no statistically significant association between prudent diet adherence and attainment of the low-density lipoprotein cholesterol target, nor with the simultaneous achievement of all three metabolic targets at once. This latter finding is particularly striking, because combined target attainment is the outcome most tightly linked to long-term complication prevention in clinical practice. The authors also conducted a series of sensitivity analyses—alternative analytical specifications designed to test whether the results were robust to different modeling choices. Across these analyses, the association with blood pressure target attainment proved the most consistent, holding up under every specification tested. The association with glycated hemoglobin remained directionally positive throughout but varied in magnitude and statistical significance depending on the analytical approach, while the findings for cholesterol and combined target attainment remained nonsignificant across the board.

Why might a prudent dietary pattern move the needle on blood pressure and glycemia but not on LDL cholesterol? Several physiological mechanisms offer plausible explanations. Diets rich in vegetables, fruits, and whole grains are typically high in potassium, magnesium, fiber, and antioxidants, and low in sodium—a profile closely aligned with the Dietary Approaches to Stop Hypertension, or DASH, diet, whose blood-pressure-lowering effects have been demonstrated in numerous randomized trials. The fiber and low glycemic load of prudent diets also blunt post-meal glucose excursions and can improve insulin sensitivity over time, which would be consistent with the observed association with HbA1c, a marker of average blood glucose over the preceding two to three months. LDL cholesterol, by contrast, is heavily influenced by saturated fat intake, body weight, genetics, and—critically in a diabetic population—statin therapy, which many participants were likely taking. When medication already drives cholesterol toward target, the marginal contribution of diet may be harder to detect in observational data.

The study’s design imposes important caveats. As a cross-sectional analysis, it captures diet and metabolic status at a single point in time, so it cannot establish whether prudent eating caused better metabolic control or whether people with better-controlled diabetes are simply more likely to eat well—a classic reverse-causation dilemma. People whose diabetes is well managed may receive more intensive nutrition counseling, have more health-conscious lifestyles overall, or possess the resources and motivation to maintain healthier diets. The reliance on a food frequency questionnaire also introduces the possibility of recall bias, since participants may not accurately remember or report their habitual intake. Furthermore, the findings come from a specific population in Guangdong Province, where Cantonese dietary traditions—characterized by rice, fresh vegetables, seafood, and soups—differ from diets elsewhere in China and around the world, limiting how directly the results can be extrapolated.

Nevertheless, the research adds a meaningful piece to a growing body of evidence that whole-diet patterns, not isolated nutrients, are the most useful lens for understanding how food shapes metabolic health in diabetes. For clinicians, the message is that dietary counseling aimed at shifting patients toward prudent eating patterns may offer measurable benefits, particularly for blood pressure, which remains one of the most powerful and under-treated risk factors for diabetic complications. For researchers, the study highlights the value of data-driven approaches such as PCA, which let the actual eating habits of a population reveal their own structure rather than forcing dietary data into predefined categories. And for the 1,465 Guangdong residents whose data formed the backbone of the analysis, the findings suggest that the everyday choices made at the dinner table may quietly shape the numbers that determine their long-term health. Future longitudinal and intervention studies, ideally across diverse dietary cultures, will be needed to confirm whether the prudent pattern can truly deliver on this promise—and to determine whether the cholesterol target, the most stubborn of the three, demands a different strategy altogether.

Subject of Research: Association between dietary patterns and metabolic control in adults with diabetes

Article Title: Association between PCA-derived dietary pattern and metabolic control among adults with diabetes in Guangdong Province: a cross-sectional study

Article References: Chen, X., Deng, H., Liu, Z., Jiang, W., Zhou, F., Xu, W., & Zeng, L. (2026). Association between PCA-derived dietary pattern and metabolic control among adults with diabetes in Guangdong Province: a cross-sectional study. BMC Endocrine Disorders. https://doi.org/10.1186/s12902-026-02501-w

Image Credits: AI Generated

DOI: 10.1186/s12902-026-02501-w

Keywords: diabetes, dietary patterns, principal component analysis, blood pressure, HbA1c, LDL cholesterol, metabolic control, Guangdong, cross-sectional study, prudent diet, nutrition, epidemiology