Exploring Vitamin D Trends Through Big Data Analysis

Big data analysis has revolutionized medical research, making it possible to analyze vast amounts of data and gain valuable insights that were previously impossible to obtain. Our knowledge of the characteristics of vitamin D sufficiency is primarily based on data from a limited number of observations, generally spanning a few years at most.

Here at the Medical Faculty of the University of Debrecen, the big data approach has allowed us to analyze trends in vitamin D status using nearly 60,000 25-hydroxyvitamin D (25(OH)D) concentration results from 2000 onwards.

Apart from analyzing the well-known phenomenon of seasonality in 25(OH)D concentration, we observed a trend in test requests, which increased from a few hundred in 2000 to almost 10,000 in 2020. Of particular interest is the change in the gender gap in test requests. In previous years, test requests were primarily from women, but by the end of the analysis period, a significant number of requests were from men as well. Since the data set includes all age groups, we analyzed 25(OH)D concentration for incremental age sets of five years, from a few months to 100 years old. The prevalence of vitamin D insufficiency (<75 nmol/L) was clearly demarcated among various years of observation, age groups, sexes, and seasons. Our data was particularly valuable for analyzing the effect of the methodology used for 25(OH)D determination. Three different methodologies were used during the study period, and clear, statistically significant bias was observed.

Our results clearly demonstrate the effect of the methodology used to determine 25(OH)D concentrations on vitamin D status, explicitly highlighting the urgent need to standardize the various platforms used to measure this important analyte and its consequences for public health.