Studies on the impact of vitamin D deficiency on health disparities are difficult to design and conduct. The word “health”refers to a wide range of diseases and conditions, but research studies typically only obtain data on one disease or on a set of related diseases or conditions.
In the case of vitamin D research, if the data doesn’t show an association between vitamin D and the health outcome in a single disease, then the conclusion by various media outlets is likely to be that vitamin D has no effect on health outcomes in any disease or any combination of diseases. Needless to say, this is obviously an unjustified assumption.
A second major issue is that vitamin D deficiency is just one determinant of health disparities. My own research suggests that socioeconomic disparities have a larger impact on health than vitamin D disparities. Yet, the impact of vitamin D on health is important because treating vitamin D deficiency is easy and cheap. There are few examples of successful treatment of health disparities resulting from improved socioeconomic status or less discrimination.
Finally, in public data sets that include measures of socioeconomic status or discrimination, there are typically no measurements of skin color or of serum 25-hydroxyvitamin D (25(OH)D), the biomarker of vitamin D status.
This lack of data is dealt with by using proxies – often race/ethnicity for skin color and environmental UV availability for vitamin D status. Skin color is a more accurate predictor of vitamin D status than race/ethnicity in this context because the reason for differences in vitamin D levels between races is that darker skin produces less vitamin D for a given level of UVB exposure than lighter skin.[private]
The difficulty in obtaining an accurate measurement of vitamin D status is seen in a recent study on race-based disparities in the rates of preterm births and low birth rates in the U.S:
The paper begins by defining the UVD hypothesis: racial disparities in health outcomes are correlated with UV light availability.
For this study, Dr. Zaneta Thayer from University of Colorado Denver gathered data from 2007 to analyze UV light availability, rates of low-weight and preterm births, and socioeconomic data, by race and U.S. state. The District of Columbia was included as a state, Alaska and Hawaii were excluded as geographic outliers, and Connecticut was excluded because of a lack of data, resulting in a sample size of 48 states.
Dr. Thayer wanted to know if disparities in rates of low-weight and preterm births were correlated with UV light availability, as the UVD hypothesis predicts.
Although randomized controlled trials, meta-analyses, and other studies have all found relationships between low maternal vitamin D status and increased preterm and low-weight infants, this study showed that states in the north with lower ambient UV had healthier infants. Moreover, disparities in rates of low-weight and preterm births were highest where UV was highest and lowest where UV was lowest.
Here is a graph from the publication showing the percentage of low birth weight infants on the left and the percentage of preterm births on the right, for a given UV index score (higher is brighter), by race:
There is a correlation in the data between southern states with higher levels of UV and increased disparities. These results do not support the UVD hypothesis.
Dr. Thayer stated that the results may be confounded by economic and social variables. In particular, even after controlling for UV, the Gini coefficients (a measure of income inequality) remained a significant positive predictor of disparities in birth weight and preterm births.
She also stated that another possible confounding factor was heat stress, since higher UV levels were strongly correlated with higher temperatures.
In any case, it’s good to see that researchers continue to study the relationship between vitamin D levels and health disparities due to skin color.[/private]