The volume of vitamin D testing in labs has risen rapidly in recent years as public awareness has grown of its complex role in maintaining health. However, the evidence on the link between vitamin D and diseases like diabetes and cancer has been inconsistent. This issue of Strategies explores a study that may help offer an explanation.

Physicians now routinely order tests for circulating 25-hydroxyvitamin D (25(OH)D) in patients at risk for osteoporosis or other bone diseases. Research has often shown a high prevalence of vitamin D deficiency worldwide. In the U.S., the National Health and Nutrition Examination Survey (NHANES) conducted by the Centers for Disease Control and Prevention from 2001 to 2004 found that more than three-quarters of U.S. adults were insufficient or deficient.

Studies have also hinted that the effects of vitamin D insufficiency may reach well beyond just bone health. Conditions as disparate as colon and breast cancer, hypertension, coronary artery disease, diabetes, and even obesity have been linked to levels of vitamin D. Now, researchers from the National Institutes of Health (NIH) say there may be good or bad times to test vitamin D, potentially elucidating some of the inconsistencies in studies that often relied on single 25(OH)D measurements (J Clin Endocrinol Metab 2013;98:97–104).

In a prospective, nationwide study, the researchers took two blood samples from 538 participants at different times of the year, examining inter- and intra-individual variability. Participants came from the U.S. Radiologic Technologist Study. The authors found that a single blood sample obtained in the spring or fall provided a decent average of a person’s circulating 25(OH)D over a 1-year period. Summer and winter samples tended to skew results.

The study could have implications for researchers exploring the connection between vitamin D and disease, according to first author, Jacqueline Major, PhD, a cancer epidemiologist who led the study from the Division of Cancer Epidemiology and Genetics at the National Cancer Institute. “This is a particularly important question,” she said. “A single measurement may not accurately reflect vitamin D levels over time, but many population-based studies are relying on a single measurement of circulating vitamin D to examine vitamin D relationships. While we found that a single measurement in the spring or fall provided a reasonable average for a one-year period, if you’re using a single measure in the winter or summer, that’s where measurement error could be introduced, and population-based study findings could be influenced by up to 40 percent, according to our findings.”

The study found that vitamin D status did not change much from the first sample to the second for most participants, with 63.6% remaining sufficient (≥50 nmol/L) and 17.5% remaining insufficient. Only 6.7% went from sufficient to insufficient, and 12.3% went from insufficient to sufficient. In addition, the correlation between the two 25(OH)D measurements was 0.75.

The public, and clinicians, often receive conflicting advice about vitamin D, which in turn can drive testing. Most recently, the U.S. Preventive Services Task Force (USPSTF) recommended against decades-old wisdom on vitamin D and calcium supplementation. In a February 2013 report, USPSTF recommended against supplementation with 400 IU or less of vitamin D3 and 1,000 mg of calcium carbonate to prevent fractures for postmenopausal women. The panel found insufficient evidence to assess the benefits of higher levels of supplementation, or any level of supplementation for premenopausal women or men.

The recommendations of a 2010 Institute of Medicine (IOM) report that established current estimated requirements for vitamin D also struck a cautious tone, warning that too much supplementation could actually be harmful. The authors of the IOM report, Dietary Reference Intakes for Calcium and Vitamin D, also judged that most U.S. adults receive adequate amounts of vitamin D in their diet.

IOM also struggled with laboratory testing for vitamin D. The cutpoints that laboratories use to report vitamin D test results have not been based on rigorous scientific studies and are not standardized, IOM found, and this lack of agreement could mean that the same individual could be declared deficient or sufficient depending on which laboratory performs the test.

In contrast to many previous studies, Major and her colleagues examined within-subject variability of 25(OH)D using the same DiaSorin LIAISON assay at the same lab. They noted that previously, “substantial variation has been reported between vitamin D assay methods, batches, and laboratories, including assay drift over long periods of time as experienced with NHANES.” Another significant difference in the NIH study was the number of participants, making it the largest study of 25(OH)D across a 1-year period in the U.S. at the time.

In addition to the size of the study, the diversity of participants makes it important as well, Major noted. “The need to reexamine this question has been around for some time. I thought that this particular study population would be able to accommodate us well, and help us examine some of the problems that previous studies have encountered when looking at these associations,” she said. “One strength of our study was that the participants were racially diverse and from across the U.S. with residents of latitudes ranging from the far north to the far south.”

In fact, the researchers examined not only demographic variables, but also environmental variables known to have an impact on vitamin D, such as exposure to the sun. They used estimates of ground level ultraviolet radiation (UVR) from the National Aeronautics and Space Administration’s Total Ozone Mapping Spectrometer (TOMS), linking each study participant’s home address to the TOMS database and estimating mean potential ambient UVR exposure for each person during the months of blood collection.

However, Major emphasized that clinical laboratories should be cautious about making conclusions based on the study’s findings. “I would note that the study was not conducted in a CLIA lab, which limits the generalizability of the findings for clinical practice,” Major said. “In addition, further studies are needed to estimate the intra-individual variability and agreement in 25(OH)D concentrations over a longer period of time than one year, and in populations that include younger adults.”

A. Catharine Ross, PhD, a professor in the department of nutritional sciences at Penn State and the chair of the 2010 IOM report on vitamin D, commended the NIH researchers for exploring variation in circulating 25(OH)D. “This type of study is important for quality control and understanding natural causes of variation and laboratory reproducibility,” Ross said. “The latter is important as guidance in that it might help to avoid an excessive number of measurements, which are expensive to the healthcare system and possibly don’t provide any useful information. A better understanding of seasonality and other variables should be useful.