Vitamin D and health in adults in Australia and New Zealand: a position statement

Vitamin D status has emerged as a significant public health issue in Australia and New Zealand. An estimated 31% of adults in Australia have inadequate vitamin D status (serum 25-hydroxyvitamin D [25-OHD] level < 50 nmol/L), increasing to more than 50% in women during winter–spring and in people residing in southern states.1 This field of research is fast-moving, and it is timely to re-examine past recommendations2 in light of the increasing number of medical conditions associated with low vitamin D status, and indications that higher levels of circulating serum 25-OHD may be required for optimal health. This article provides updated guidance to clinicians and health professionals on the role of vitamin D in health for adults. The guideline development process is outlined in Box 1.

The generic term “vitamin D” is generally applied to two molecules: cholecalciferol and ergocalciferol. Cholecalciferol (vitamin D3) is formed through the action of ultraviolet B (UVB) radiation (wavelength, 290–315 nm) on 7-dehydrocholesterol in the skin. This process is the major determinant of vitamin D status, as most adults are unlikely to obtain more than 5%–10% of their vitamin D requirement from dietary sources. Cholecalciferol is also the major form of supplemental vitamin D currently available in Australasia. The other form of vitamin D, ergocalciferol (vitamin D2), is produced by UV irradiation of the plant sterol ergosterol. Few vitamin D supplements now contain this form. Both forms of vitamin D are transported to the liver and metabolised to 25-OHD. This is the major circulating form and the metabolite routinely used to assess overall vitamin D status. It is also likely to be the metabolite responsible for hypercalcaemia when vitamin D is given in excess.3

Further hydroxylation occurs in the kidney to form the biologically active 1,25-dihydroxyvitamin D (1,25-(OH)2D), also known as calcitriol. This compound promotes absorption of calcium and phosphate from the gut, contributes to extracellular calcium and phosphate homeostasis directly and through an interaction with parathyroid hormone (PTH), facilitates mineralisation of the skeleton, and is important for muscle function.4 In addition, almost every nucleated cell expresses the vitamin D receptor, and many extrarenal tissues have the capacity to make 1,25-(OH)2D. Thus, vitamin D can operate via classical endocrine pathways (renal synthesis of 1,25-(OH)2D under the control of calcium concentration via PTH, or phosphate, with secretion into the bloodstream to increase intestinal calcium absorption), as well as through autocrine and paracrine pathways involving local synthesis and actions.

For most people, the main source of vitamin D is skin exposure to sunlight. The minimal erythemal dose (MED) is the amount of UV radiation exposure that just causes faint redness of the skin (erythema). At low wavelengths, in the UVB range, the action spectrum for production of erythema is similar to that for production of vitamin D. However, erythema, but not vitamin D, can also be produced by higher-wavelength UVA radiation. This limits the usefulness of MED as an index of vitamin D dose.5 Nevertheless, there are experimental data indicating that exposure of around 15% of the body surface (arms and hands or equivalent) to one-third of an MED near the middle of the day will result in the production of about 1000 IU (25 μg) of vitamin D. Achieving this exposure on most days should generally, though not always, be sufficient to maintain vitamin D levels in the body.5,6

The amount of sun exposure required to produce one-third of an MED depends on several factors (Box 2). Less vitamin D is synthesised in winter, particularly at latitudes further from the equator. There is minimal transmission of UVB radiation through normal window glass. Although sunscreens also markedly reduce transmission in the UVB range, these are often inadequately applied and may have little impact on vitamin D status.7 Lack of any skin exposure to sunlight, through confinement indoors or from clothing, is a more common issue. Short UV radiation exposures (of a few minutes) may be more efficient at producing vitamin D, as prolonged exposure to UV radiation results in the production of sterols that have little effect on mineral metabolism.4

For people with moderately fair skin, adequate vitamin D levels are likely to be maintained in summer by a walk outside with arms exposed for 6–7 minutes mid morning or mid afternoon, on most days. In winter the task is more difficult but, depending on latitude, walking outside at lunchtime for 7–40 minutes, with as much bare skin exposed as feasible, on most days, is likely to be helpful although at very southern latitudes, relatively little vitamin D would be formed5(Box 2). Care needs to be taken by people in the far north of Australia where the UV intensity remains moderate to high throughout the year. Given the high incidence of skin cancer in Australasia, sunscreens and other UV radiation avoidance measures should be used if exposure is likely to be prolonged when the UV index is 3 or above and/or there is a risk of skin damage, even for people with vitamin D deficiency. Some people are at particularly high risk of skin cancer, including those with a past history of skin cancer and those who are immunosuppressed. Such people need more rigorous sun protection and should discuss with their medical practitioner whether supplements might be more appropriate than sun exposure to maintain vitamin D levels.

Adult groups at high risk of vitamin D deficiency are shown in Box 3. Although older people are an at-risk group for developing vitamin D deficiency, because they have less substrate (7-dehydrocholesterol in the skin) and lower production rates of vitamin D from high levels of UV radiation exposure,13 they synthesise similar amounts of vitamin D as younger people at normal levels of UV radiation exposure.14 A major reason for vitamin D deficiency in older people is limited sun exposure.9

Vitamin D synthesis is reduced in those who have dark skin due to the presence of melanin, which absorbs UV radiation.7 People with dark skin are likely to need sun exposures 3–6 times longer than the times shown in Box 2. Vitamin D synthesis is also reduced by clothing,7,15,16 and people who wear covering clothing for cultural or religious reasons are an at-risk group. The sun avoidance behaviour of fair-skinned people increases their risk of vitamin D deficiency.11,17 Reduced sun exposure is also likely to contribute to increased risk of vitamin D deficiency in people with chronic illness and those confined indoors. There is some evidence that sun exposure and vitamin D status may have separate roles in the development of disease.18

Vitamin D3 is found naturally in small quantities in a few foods, such as wild-caught fatty fish (eg, North Sea salmon, herring, mackerel). Liver, eggs and fortified foods such as margarine and some low-fat milk products also contain very small amounts of vitamin D3. In Australia, there is one specially supplemented milk product with 200 IU (5 μg) of vitamin D per 250 mL serve, and some mushrooms that have been exposed to UV radiation provide 800 IU (20 μg) of vitamin D2 per 100 g.4 However, for most people, dietary vitamin D intake is limited.

Although accurate, comprehensive data on the vitamin D content of foods in Australasia are not available, vitamin D intake from dietary sources is likely to be insufficient to meet requirements, as the average estimated daily dietary intake for adults is only between 80 and 120 IU (2–3 μg).19 This is considerably lower than in other countries (eg, Canada and the United States, where average intake ranges between 120 and 240 IU [3–6 μg] per day) where more extensive vitamin D fortification of the food supply is mandated or permitted.20 It is acknowledged that the current (2006) guidelines for recommended dietary intakes (ie, adequate intakes) of vitamin D in Australia and New Zealand21 are out of date. The recently revised recommended daily allowances (RDAs) for vitamin D in the US are 600 IU (15 μg) for people aged 1–70 years and 800 IU (20 μg) for those aged ≥ 71 years, with an upper limit (that includes a generous safety factor) of 4000 IU (100 μg).22

The most clearly established effects of vitamin D are to maintain calcium and phosphate homeostasis, and to optimise bone health and muscle function.23-25 The hormonal form, 1,25-(OH)2D, increases active intestinal calcium (and phosphate) absorption, which helps offset obligatory calcium losses from the kidneys, gut and skin. Calcium concentrations in blood are maintained within narrow limits, so if inadequate calcium is absorbed from dietary sources, bone resorption increases under the influence of PTH, to maintain circulating calcium levels. Severe vitamin D deficiency causes impaired bone mineralisation, resulting in rickets in children and osteomalacia in adults. There is Level II evidence that optimal mineral metabolism, bone density and muscle function is achieved at serum 25-OHD concentrations of 50–60 nmol/L,23-26 with no consistent evidence that higher levels are beneficial. Although most of these studies have been in older people, there is some evidence of improvement in bone mineral content from vitamin D supplementation in children and adolescents aged between 1 month and 20 years with low vitamin D levels (< 35 nmol/L).27 Further, a recent histomorphometric study of 675 German adults showed increased osteoid (unmineralised bone matrix) in some people with 25-OHD concentrations < 75 nmol/L, suggesting that this may be an appropriate cut-off for optimal bone health.28

Vitamin D deficiency is an independent predictor of falls in older people, and circulating 25-OHD levels < 60–75 nmol/L have been associated with lower-extremity muscle weakness and impaired balance, and accelerated losses in muscle mass, strength and physical function.25,29 A recent meta-analysis indicated that a minimum serum 25-OHD level of 60 nmol/L is required to effectively reduce the rate of falls,30 although this analysis has been disputed.22 Most Level I evidence indicates that vitamin D (at daily doses of > 800 IU [20 μg]) needs to be combined with adequate calcium (> 1000 mg per day), rather than vitamin D alone, to reduce the risk of falls and fractures; although there may be benefits with single therapies.30-34 Therefore, older people would be recommended to consume adequate (1000–1300 mg per day) but not excessive dietary intakes of calcium, together with maintaining adequate vitamin D status (within the 25-OHD range 50–60 nmol/L, allowing for seasonal variation), to reduce risk of falls and fracture. This recommendation is consistent with the recent review conducted by the US Institute of Medicine that recommends a minimum serum level of 50 nmol/L, which can be achieved by meeting the RDA for vitamin D.22,35

A wide range of diseases have been associated with low levels of circulating serum 25-OHD,4,13 including autoimmune diseases,36,37 cardiovascular and metabolic diseases,4,38-40 some cancers,41,42 microbial and respiratory diseases,4,43-45 and some neurological and mental health conditions46-48 including schizophrenia,49 as well as all-cause and cardiovascular mortality50,51 (Box 4). Most of these studies were observational and did not adjust for important confounders. While there are studies in animal models that support the epidemiological evidence, and plausible mechanisms to explain the effects of vitamin D, there are very few randomised controlled trials (RCTs), most of which have marked limitations, on which to base recommendations.39,41,44 Accordingly, proposals that recommend serum 25-OHD levels of 75–80 nmol/L or higher56 are not supported by a significant amount of data from RCTs.22

Other epidemiological studies have recently provided Level III evidence of U-shaped exposure–risk relationships between serum 25-OHD levels and disease outcomes (ie, increased risk at both low [< 30 nmol/L] and high [> 75 or > 125 nmol/L] concentrations) for mortality,57,58 schizophrenia,49 prostate cancer,59 and frailty in women.60 These U-shaped curves may in part reflect a mixture of common genetic variants,61 but further RCTs are needed.

Vitamin D enters adipose tissue and may not be readily released, unless there is fat breakdown, so obesity results in lower vitamin D levels after receipt of oral or cutaneously synthesised vitamin D.12 Physical activity is a predictor of vitamin D status,62 which could be due to alteration in the storage and mobilisation of vitamin D in fat63 or to greater exposure to sunlight when exercising. Polymorphisms in the genes for vitamin D-binding protein, 7-dehydrocholesterol reductase (which affects the amount of substrate 7-dehydrocholesterol in skin) and 25-hydroxylase may contribute as much to variation in 25-OHD levels as the summer–winter difference.61,64 The half-life of 25-OHD is long and, for reasons that are not entirely known, varies between 15 and 50 days.65 Some medications, such as cytochrome P450 enzyme inducers, including some anticonvulsants, accelerate degradation of vitamin D compounds.66

One important factor is calcium intake, with low intakes of calcium or high levels of PTH (either primary or secondary) linked to accelerated degradation of 25-OHD and a shorter half-life.67 In contrast, increased dietary calcium intake leads to higher 25-OHD concentrations.68,69 Low calcium absorption may be the reason why people with malabsorption, or who undergo gastrectomy or related procedures,69 have reduced 25-OHD levels, as there is no evidence for clinically significant enterohepatic circulation of vitamin D compounds.70

The main concerns with excessive vitamin D levels are hypercalciuria and hypercalcaemia. Vitamin D toxicity can be caused by excess oral intake through supplementation, but not by prolonged exposure of the skin to sunlight, which produces 25-OHD values in the range 150–200 nmol/L.4,5,71 Hypercalcaemia is not seen until serum 25-OHD levels reach 220 nmol/L and is generally not reported until levels reach 500 nmol/L.71 There is no evidence of toxicity, based on blood calcium concentrations, at vitamin D doses up to 5000 IU per day or 50 000 IU per month.72-74

However, adverse effects can probably no longer be defined solely by hypercalcaemia, as a large RCT in older Australian women found that a single annual dose of 500 000 IU vitamin D3 for 3–5 years resulted in a 15% increase in falls and 26% increase in fractures.75 The increased risk of falls was pronounced in the first 3 months after taking the dose, and fracture risk also tended to be higher during this 3-month post-dose period, when serum 25-OHD levels would have been highest. Annual megadose vitamin D is thus not recommended, although a 5-year community study providing 100 000 IU (2500 μg) of oral vitamin D3 every 4 months to older people did show benefits for fracture reduction.76 Furthermore, intramuscular administration of vitamin D (100 000 or 150 000 IU) may be appropriate in patients with malabsorption, although intramuscular preparations are not readily available in Australia. Administration of vitamin D is contraindicated in most cases of hypercalciuria or hypercalcaemia. Vitamin D treatment is not contraindicated in patients with primary hyperparathyroidism and vitamin D deficiency.77 Cod liver oil, which can be used for vitamin D supplementation, also contains vitamin A, which can be toxic at high doses.

Based on our review of the available evidence, vitamin D status can be defined according to the following levels of serum 25-OHD:

• • Vitamin D adequacy: ≥ 50 nmol/L at the end of winter (level may need to be 10–20 nmol/L higher at the end of summer, to allow for seasonal decrease).

  • Mild vitamin D deficiency: 30–49 nmol/L

• Moderate vitamin deficiency: 12.5–29 nmol/L

• Severe vitamin D deficiency: < 12.5 nmol/L

On the basis of the evidence discussed above, a target level for vitamin D adequacy for mineral homeostasis, bone health and muscle function would seem to be > 50 or 60 nmol/L,22,24 although optimal values, even for bone and muscle health, are not clear. Ideally, the target value should allow for a decrease in vitamin D levels during winter.17 An international consensus statement recently agreed on a 25-OHD serum concentration range between 50 and 62.5 nmol/L (20–25 ng/mL) to prevent adverse musculoskeletal outcomes, including falls and fractures.78 The target serum level for prevention of other diseases is not clear but may be higher, in the range of 75–80 nmol/L (30–32 ng/mL). However, there is currently limited high-level evidence to support these higher 25-OHD serum levels.56

Clinicians, researchers and policymakers should be aware of the imprecision of current 25-OHD testing, and exercise caution when interpreting results in clinical practice. Although the performance of radioimmunoassay and enzyme-linked assays is acceptable, the bias and imprecision of many automated methods may be problematic at the lower, clinically and analytically important range (< 50 nmol/L) of the assay. While adoption and alignment of assays to the National Institute of Standards and Technology reference material should reduce bias,79 imprecision will remain problematic. Consequently, some laboratories are using more exacting methods of analysis, such as liquid chromatography–tandem mass spectrometry.79 All Australian and New Zealand laboratories offering 25-OHD testing are required to be enrolled in external proficiency programs (such as the Vitamin D External Quality Assessment Scheme [DEQAS]), which allow each laboratory to monitor its performance compared with its peers.

Some studies have shown that with supplemental vitamin D3, 25-OHD levels reach a plateau at about 8 weeks, after rising by, on average, about 0.7 nmol/L for each 40 IU (1 μg) per day of supplemental vitamin D3 (ie, an increase of about 17 nmol/L for each 1000 IU [25 μg] per day).73,80 However, at any given dose there is considerable individual variation in the plateau level achieved. Although the increment is negatively correlated to the baseline 25-OHD concentration, it is not significantly affected by race, sex or age.80,81 Obesity, however, is associated with lower levels of 25-OHD and a reduced dose response to oral vitamin D or UV radiation.7,12 As it may take up to 2–5 months for serum levels of 25-OHD to plateau,80,82 retesting should not take place before 3 months.

There appears to be little difference in 25-OHD values whether the total amount of vitamin D is given as a daily, weekly or monthly dose, with no clear indication that one dosing schedule is more effective than another. Intermittent dosing may be more convenient in some settings, such as nursing homes,83 and may also achieve better long-term compliance.73 There is conflicting evidence about whether vitamin D3 is more effective than vitamin D2 in raising 25-OHD levels,84,85 but virtually all oral vitamin D supplements available in Australia and New Zealand are vitamin D3. There are currently no generally available, Therapeutic Goods Administration-approved parenteral preparations of vitamin D, although these might be helpful where ability to swallow, malabsorption or compliance are problems.

Recommendations for assessment and management of vitamin D deficiency are shown in Box 5. To treat moderate to severe deficiency, it would be reasonable to use 3000–5000 IU (75–125 μg) of vitamin D per day for at least 6–12 weeks, although there are no long-term safety data for these doses. This will usually return serum 25-OHD levels to the target range, although smaller daily doses of 2000 IU might be expected to increase 25-OHD levels by 34 nmol/L, as noted earlier. As there is such an individual variation of response to vitamin D supplementation, 25-OHD levels should be checked after 3 months. If adequate levels are not achieved, consideration should be given to possible underlying gastrointestinal disorders, such as coeliac disease. Most patients whose 25-OHD levels have normalised will need ongoing treatment at a maintenance dose of around 1000–2000 IU (25–50 μg) per day or equivalent.

For adults in disadvantaged communities at high risk of vitamin D deficiency, such as dark-skinned migrants from low socioeconomic backgrounds, a strong case could be made to routinely provide vitamin D supplements, at the same dose used to treat moderate to severe deficiency, as early as possible, without initially measuring serum 25-OHD levels. As low calcium intake and associated hyperparathyroidism increase the degradation of vitamin D compounds,67,69,84 a daily intake of 1000–1300 mg of calcium, preferably from calcium-rich foods, should be encouraged.

According to the best evidence currently available, adequate serum 25-OHD levels of ≥ 50 nmol/L at the end of winter (10–20 nmol/L higher at the end of summer to allow for a seasonal decrease), along with recommended dietary calcium intakes and weight-bearing or muscle strengthening exercise, are required for optimal bone and muscle function, at least in older age groups. Although it is likely that serum 25-OHD levels that are somewhat higher than those required for musculoskeletal health play a small role in prevention of some disease states, there is evidence emerging of potential adverse effects of higher serum 25-OHD levels under some circumstances. RCTs to test these hypotheses need to be undertaken before higher targets can be recommended with confidence.