Measuring exposure to vitamin D is relatively easy -- it's a simple matter of serum levels of a compound called 25-hydroxyvitamin D, or 25(OH)D, Rosen said.
But the relationship between circulating 25(OH)D and the active form of the vitamin, 1,25-dihydroxycholecalciferol, is not clear. It's entirely possible, he said, to have low levels of 25(OH)D and yet have a perfectly adequate amount of the hormonally active form.
Indeed, Manson said, the Institute of Medicine (IOM) recently estimated that the average requirement of 25(OH)D is really only 16 nanograms per milliliter -- a level that would in most cases ensure adequate amounts of the active vitamin.
"It's extremely variable," she noted, "and there is much that isn't known about vitamin D and metabolism."
"So the tissues and cells may be seeing adequate amounts of biological active vitamin D and adequate stimulation of the vitamin D receptor even in those who have lower blood levels," she said.
So the question of how much vitamin D is enough is a vexing one. But doctors and their patients still want an answer.
The IOM last year released new guidelines for vitamin D, which say that healthy people should aim to have at least 20 nanograms of 25(OH)D per milliliter of serum.
That can be achieved, the IOM said, by taking 600 IU a day of a vitamin D supplement if people are between ages 1 and 70, and 800 a day if they are 70 or older.
Manson, who along with Rosen was part of the IOM panel, said that "relatively modest amount" of vitamin D will keep 97.5 percent of the general population in good bone health.
And, she noted, the IOM really was only concerned about bone health, because there's such a dearth of evidence for benefits in other areas.
But even if there's no evidence for a benefit, is there any harm in taking a bit extra? Well, possibly.
One of the functions of vitamin D is to regulate calcium and phosphorus; too much can lead to hypercalcemia. The IOM set the tolerable upper limit at between 2,500 and 4,000 IU per day, depending on age.
The institute also cautioned, Manson said, that there is some evidence of a U-shaped curve for vitamin D -- too little is bad and so is too much.
In particular, the IOM reported that, although the evidence is weak, more than 4,000 IU a day of the vitamin might increase the risk of cardiovascular disease, some cancers, and all-cause mortality.
"There's no evidence that more is better, so why use more?" Rosen asked.
On a day to day basis, most people can ensure they have enough vitamin D by eating certain foods – fatty fish like salmon, for example – and taking a multivitamin, Manson said.
But there's no need to screen the healthy population for vitamin D levels at least until there's more evidence that it matters, the IOM concluded.
On the other hand, the Endocrine Society has called for regular screening for groups at risk for vitamin D deficiency such as the obese, African Americans, and pregnant women.