Yiqiang Zhan1, Zheng Tang, Jinming Yu. 1. Institute of Clinical Epidemiology, School of Public Health, Fudan University, 130 Dong'An Road, Shanghai, People's Republic of China, zhany09@fudan.edu.cn.
Abstract
AIMS: The present study aims to investigate the association between serum ferritin and diabetes, diabetes control, and insulin resistance (IR) and examine whether gender is a modifier for these associations in a community-based sample. METHODS: A cross-sectional survey of 8,235 participants was conducted in 2009. Serum ferritin, glucose, hemoglobin A1c (HbA1c), insulin, inflammatory markers, and lipid markers were measured. IR was estimated with a Homeostasis Model Assessment (HOMA-IR) equation. Multiple logistic and linear regression models were applied to evaluate these associations. RESULTS: The numbers of diabetic patients and non-diabetic participants in the present study were 644 (7.8 %) and 7,591 (92.2 %). After adjusting for multiple confounders, the odds ratios (ORs) and 95 % confidence intervals (CIs) for diabetes were 1.48 (1.31-1.69) in men and 1.43 (1.24-1.65) in women for one-unit increase in log-transformed serum ferritin levels. Likewise, ORs (95 % CIs) for poor diabetes control (HbA1c ≥7.5 %) were 1.58 (1.21-2.05) and 1.37 (1.07-1.77) in men and women, respectively. As for HOMA-IR, the respective betas (P value) for one-unit increase in log-transformed serum ferritin were 0.07 (P < 0.0001) and 0.06 (P < 0.0001) in men and women. CONCLUSIONS: In conclusion, elevated serum ferritin levels were associated with higher risks of diabetes, higher levels of HbA1c, and HOMA-IR independent of several confounders.
AIMS: The present study aims to investigate the association between serum ferritin and diabetes, diabetes control, and insulin resistance (IR) and examine whether gender is a modifier for these associations in a community-based sample. METHODS: A cross-sectional survey of 8,235 participants was conducted in 2009. Serum ferritin, glucose, hemoglobin A1c (HbA1c), insulin, inflammatory markers, and lipid markers were measured. IR was estimated with a Homeostasis Model Assessment (HOMA-IR) equation. Multiple logistic and linear regression models were applied to evaluate these associations. RESULTS: The numbers of diabeticpatients and non-diabeticparticipants in the present study were 644 (7.8 %) and 7,591 (92.2 %). After adjusting for multiple confounders, the odds ratios (ORs) and 95 % confidence intervals (CIs) for diabetes were 1.48 (1.31-1.69) in men and 1.43 (1.24-1.65) in women for one-unit increase in log-transformed serum ferritin levels. Likewise, ORs (95 % CIs) for poor diabetes control (HbA1c ≥7.5 %) were 1.58 (1.21-2.05) and 1.37 (1.07-1.77) in men and women, respectively. As for HOMA-IR, the respective betas (P value) for one-unit increase in log-transformed serum ferritin were 0.07 (P < 0.0001) and 0.06 (P < 0.0001) in men and women. CONCLUSIONS: In conclusion, elevated serum ferritin levels were associated with higher risks of diabetes, higher levels of HbA1c, and HOMA-IR independent of several confounders.