The contraceptive pill and vitamins: which needs do studies describe, and for which nutrients?

Studies measure lower blood levels of several vitamins in pill users, but they do not establish that the pill creates deficiencies in women whose diet is adequate.

A frequently cited review on the subject, published in 2013 by a team from Sapienza University of Rome, goes through the nutrients for which differences have been reported: folic acid, vitamins B2, B6 and B12, vitamins C and E, magnesium, selenium and zinc¹. Its authors conclude that clinicians should consider supplementation as a first-line measure. It is a narrative review, with no systematic protocol for selecting studies and no formal assessment of their quality, and its conclusion goes further than the research it brings together.

An older Dutch trial reached the opposite conclusion. For four cycles, 59 women, 28 of them on a pill containing 30 µg of ethinylestradiol and 31 controls, received a multivitamin supplement with folic acid after one cycle of baseline measurements. Vitamin A was higher and vitamin B12 lower in pill users, with no baseline difference in riboflavin, vitamin C or blood and red cell folate. Supplementation raised the same vitamins in both groups, and the authors considered it unjustified in healthy young women, while regarding it as an option where vitamin balance is fragile or folate needs are higher².

Vitamin B6 and vitamin B12 are the two for which measurements agree best. Folate and riboflavin give far more disputed results.

For vitamin B6, the US NHANES 2003-2004 survey measured plasma pyridoxal 5’-phosphate, the marker used to set recommended intakes, in more than 6,000 people. Women of childbearing age had significantly lower concentrations than men of the same age, whether or not they took a supplement, and most oral contraceptive users were below 20 nmol/L, the threshold the authors used for a low level³. An Italian study of 219 women, 60 of them on the pill, found a median vitamin B6 level of 24.2 nmol/L in users versus 32.9 nmol/L in the others⁴.

For vitamin B12, the same Italian study reports a lower median in users (278 versus 429 ng/mL), with no difference in homocysteine or folate⁴. A German study compared 71 women who had been taking a pill containing 20 µg of ethinylestradiol for at least three months with 170 controls: cobalamin was lower in users, nine of them had clearly low cobalamin without any symptoms, and folate did not differ. The authors conclude that routine testing is not justified, and that a supplement is rather a matter for discussion in women who are already deficient or who follow restrictive eating habits⁵.

The question is not settled: a 2015 meta-analysis measured a fall, whereas a 2011 review judged that the data do not demonstrate one for current pills.

A Canadian meta-analysis published in 2015 pooled 17 studies and 2,831 women. In users, it measured an average fall in plasma folate of 1.27 µg/L, and a fall in red cell folate of 59.32 µg/L across 12 studies and 1,389 women. Its authors conclude that it is essential to continue folate supplementation during oral contraception⁶.

A 2011 US review, devoted specifically to folate, B6 and B12 during oral contraception, comes to a different reading. Its authors point out that most of the studies that concluded there was a negative effect on folate were carried out when pills contained much more oestrogen, and that many did not control for confounding factors. In their view, the available data do not allow the conclusion that current pills worsen folate status. They do, however, retain an effect on vitamin B6, and consider that functional markers of B12 status are not significantly affected⁷.

These two conclusions coexist in the literature. They rest on studies that differ in diet, flour fortification from one country to another, duration of use and supplement intake, and these factors alone can shift the averages.

British research from the early 1970s showed that the combined pill alters a metabolic pathway that depends on vitamin B6: the tryptophan pathway.

In a study published in 1972 in 10 women on a combined pill containing 50 µg of oestrogen, urinary excretion of xanthurenic acid rose from day 21 of pill use, and several tryptophan metabolites were raised at three months. One month of pyridoxine at 20 mg a day fully corrected this urinary profile, and the abnormality could persist for three months or more after stopping. A progestogen taken alone, megestrol acetate, did not cause it⁸. This dose of oestrogen is higher than that of the 20 or 30 µg pills studied in the other research cited here. Because vitamin B6 is involved in making serotonin from tryptophan, the hypothesis of a link with mood has circulated widely.

A Mexican trial tested this idea directly on the side effects that occur when starting the pill. One hundred and twenty-four women received, triple-blind, 150 mg of vitamin B6 a day or a placebo for 30 days, with scoring of nausea, headache, vomiting, dizziness, mood and irritability. Symptoms decreased in both groups, with no statistically significant difference, and the authors attribute this improvement to the placebo effect rather than to the vitamin⁹. A biological abnormality corrected by a nutrient therefore does not guarantee that a symptom will be.

A randomised placebo-controlled trial measured a fall in perceived vitality on the combined pill, but the fatigue reported in everyday life has many other possible causes.

In this double-blind Swedish trial, 340 women aged 18 to 35 received, for three months, a pill containing 150 µg of levonorgestrel and 30 µg of ethinylestradiol, or a placebo. The overall psychological well-being score fell by 4.12 points compared with placebo (95% confidence interval: −7.18 to −1.06). The vitality dimension fell by 6.84 points, self-control by 6.63 points and positive well-being by 3.90 points. Depressive symptoms, by contrast, did not differ significantly¹⁰.

This measurement applies to one specific combined pill, over three months, in healthy women. It says nothing about the hormonal IUD or the implant, which were not tested here. A 2011 review of placebo-controlled trials argues, on the contrary, that non-specific side effects, fatigue among them, are no more frequent on the combined pill than on an inactive tablet, and points either to the nocebo phenomenon or to the background frequency of these complaints in the population¹¹.

As for nutrients, the European regulation allows it to be stated that vitamin B6, folate, riboflavin and vitamin B12 contribute to the reduction of tiredness and fatigue, and that vitamin B6 contributes to the regulation of hormonal activity¹². These wordings describe the role of a nutrient in the body and do not promise to make fatigue that appeared on contraception disappear. Fatigue that persists deserves to be examined by a doctor, who can look for other causes, from sleep to the thyroid and iron.

For these nutrients, the data are much older and weaker than for vitamins B6 and B12.

Riboflavin was studied as early as 1974 in Thailand, using erythrocyte glutathione reductase activity as the marker, which was lower in women on the pill. The later studies cited by the same review mainly describe a deficiency in women from low-income backgrounds that the pill seemed to accentuate, which makes it hard to transpose the findings¹. In the Dutch trial, riboflavin did not differ at baseline between users and non-users, and the supplement raised it equally in both groups².

For minerals, the 2013 review brings together studies, often old, that report lower serum zinc and magnesium in users, as well as research on selenium with mixed results, one study failing to reach statistical significance¹. These measurements describe associations at a single point in time, in populations and with hormone doses very different from those of today.

For vitamin C, the studies gathered by the same review report lower concentrations in the platelets and white blood cells of users, and attribute these differences to oestrogen. Part of the gap may reflect a redistribution of the vitamin into tissues rather than a loss, and a study covering periods of six months to seven years found no threat to vitamin C status when the diet provided enough¹.

Three things remain without solid evidence: that a well-nourished woman develops a deficiency because of her pill, that a supplement reduces the side effects of contraception, and that blood levels reflect tissue stores.

The Mexican trial of vitamin B6 showed no benefit on the symptoms of starting the pill⁹. The Dutch trial did not show that supplementation acted differently in users and in other women². The German study points out that low cobalamin can be seen without the slightest clinical sign⁵, and the 2011 review notes that functional markers of B12 status do not change⁷.

Measurements agree on one point: several nutrients come out lower in the blood of pill users, with vitamin B6 first among them. Additional intake is mainly worth discussing when the diet is poor or restrictive, and when a pregnancy is being considered. A blood test and adding a supplement are decided with your doctor or pharmacist, who knows your treatment as a whole.