Discrepancies between current guidance from NICE on the treatment of vitamin D deficiency and the recommended daily amounts [RDAs] for its prevention in the UK

The National Institute for Clinical Excellence [NICE] advice for health professionals on managing vitamin D deficiency in the United Kingdom [UK] [https://cks.nice.org.uk/topics/vitamin-d-deficiency-in-adults/prescribing-information/calcium-supplements, [last updated Sept 2021], defines vitamin D deficiency by a serum 25(OH)D value of <25 nmol/l, [advised solely for bone health] and lists investigations that may be needed before advising on its treatment. In brief, treat for deficiency if 25(OH)D < 25 nmol/l in symptomatic patients and in those taking medication known to cause deficiency, preferably with vitamin D₃, in amounts totaling ‘280,000–300,000 IU over six to ten weeks,’ followed by ‘maintenance supplementation at 800 IU/day for life.’ For managing vitamin D insufficiency [serum 25(OH)D = 25–50 nmol/l], NICE advises starting maintenance dosing [800 IU/day] without any loading doses. Thus, UK adults known to have had deficiency/insufficiency [25(OH)D < 25 nmol/l and 25–50 nmol/l, respectively], should take 800 IU/day for life while those whose UK defined deficiency/insufficiency remains undetected and untreated, despite the additional non-skeletal health benefits from correction of ‘deficiency’ that are now emerging [as discussed below] that increase the need to correct undiagnosed deficiency/insufficiency.

The question, therefore, is whether deficiency/insufficiency is common enough to warrant action at the population level. In fact, deficiency of vitamin D [at serum 25(OH)D values <25(OH)D nmol/l] is present in all UK population groups summer and winter, especially in Black, south Asian and other Minority Ethnic [BAME] groups and well known to have been causing rickets and osteomalacia at higher rates in those population groups than in other Britons since first reported amongst south Asian immigrants in the 1970s, but, not being corrected, ‘… diet frequently became a proxy or shorthand for culture (and religion, and race), while disease justified pressure to assimilate’’[1].

After reconsidering National advice following continuing concerns of those knowledgeable about the mechanistic effects of vitamin D relevant to COVID-19 illness [2,3], NICE, Public Health England [PHE] and the Specialist Advisory Committee on Nutrition [SACN] issued a joint statement reaffirming existing advice but stating that the recommended daily amount [400 IU/day] should be taken all year round since COVID-19 precautions were probably limiting outdoor activity. www.nice.org.uk/guidance/ng187;17/12/2020. If followed, this advice would ensure intakes totaling 28,000 IU over ten weeks, 10% of the amount advised by NICE for correcting deficiency while maintenance treatment at 800 IU/day would provide 20% of the intake advised for correcting deficiency and would be likely to leave many people with deficiency with 25(OH)D values below the 25 nmol/l SACN recommendation for bone health.

The further question, therefore, is what proportion of the UK adult population has deficiency as defined in this country? If it is high, then, since current guidance on recommended daily amounts of vitamin D adequate to prevent 25(OH)Ds from falling below 25 nmol/l, i.e. 400 IU/day, cannot be expected to correct deficiency, an update in public health advice would be needed, together with a fresh approach to vitamin D provision at the population level, both in the UK and in other similarly affected countries.

Recent surveys found vitamin D deficiency/insufficiency [serum 25(OH)D values <25 and 25–50 nmol/l in the UK] in 30%–60% of adults across Western Europe in 2018[4], and at 15%/47%, summer/winter in the 1958 British Birth Cohort in subjects aged 45 years in 2003 [5]. Deficiency [25(OH)Ds <25 nmol/l] was present in up to 95% of south Asians in east London in the 1990s. Deficiency rates averaged 20–23% in winter in UK biobank volunteers at recruitment 10–14 years ago [and was more common in Scotland than England]. South Asian, Black and White UK-Biobank subjects at recruitment to UK-Biobank had deficiency rates of 57.2%, 38.5% and 17.5% in winter and of 50.8%, 30.8% and 5.9% in summer, respectively [6–8]. Britons, do, therefore, suffer a significant prevalence of deficiency as currently defined solely for bone health, especially BAME communities. This adult deficiency precipitates osteomalacia, worsens osteoporosis and reduces muscle strength which adds to the risks of falls and fragility fractures in the elderly, including the risks of fractured neck of femur requiring hospitalization that increase with age, increase mortality and increase the risk of becoming housebound by 40%. https://www.nice.org.uk/…/nice-impact-falls-and-fragility-fractures.

Clearly, daily intakes >400 IU/day are necessary to reduce the reported rates of deficiency in the UK. However, increased intake recommendations will be no more effective than current advice without programs that ensure adequate vitamin D intakes at the population level, best achieved by food fortification programs suited to local lifestyles. For example, from 2003 Finland has fortified milk and fatty spreads [voluntarily] while encouraging deficiency risk group supplementation; in 2010 Finland’s food fortification was doubled which has successfully minimized deficiency apart from that in recent immigrant groups [9]. Such programs also reduce 25(OH)D assay costs to those required for medical indications and for auditing fortification program efficacy.

Serum 25(OH)D thresholds for non-bony health-benefits often exceed 25 nmol/l [10] and NICE and SACN should, therefore, consider issuing revised advice for avoiding, and correcting, deficiency to reduce those problems as well as reducing musculo-skeletal problems. For example, reducing insulin resistance in deficiency in south Asians and others requires 25(OH)D values ≥80 nmol/l and, while no overall reduction in Type 2 Diabetes Mellitus [T2DM] risk was found in subjects with prediabetes in the D2d trial of supplemental vitamin D₃ at up to 4000 IU/day, T2DM risks did fall by up to 70% in D2d subjects who achieved intratrial 25(OH)Ds ≥100 nmol/l [11], an important benefit that could help combat globally rising T2DM rates and reduce the burdens that T2DM places on individuals, health services and health care budgets [12]. Notably, D2d subjects needed 4000 IU/day intakes to achieve 25(OH)D values ≥100 nmol/l, probably because hyperglycemia reduces hepatic 25-hydroxylation [13]. Over-weight and obese subjects need higher intakes [x1.5-fold and x2-3-fold, respectively], than normal weight people to achieve comparable 25(OH)Ds [14]. Clearly all such conditions should be included in the list of conditions NICE identifies as requiring higher than usual vitamin D intakes for treating and preventing deficiency, such as malabsorption. Flour fortification has been suggested for cost-effectively preventing deficiency in British adults in combination with routine [free] supplementation of those at high-risk of deficiency [those in care homes, the elderly, the house-bound, pregnant and nursing mothers, young children, indoor and shift workers, the obese, and Black, south Asian, and other ethnic minority groups] [15] but such programs would need to be acceptable to each of these disparate high-risk groups.

NICE does not support measuring 25(OH)D in asymptomatic people, leaving undiagnosed deficiency uncorrected and increasing the risks of avoidable bony deficiency disorders, a situation that would not be considered acceptable for other vitamin deficiencies [folate; B group vitamins; vitamins B12, A or C]. Abolishing vitamin D deficiency across the UK, and elsewhere, would reduce bony health risks in the long-term and be expected to provide additional significant non-bony health benefits identified from recent RCT data. For example, reducing the risks of respiratory infections with daily dosing [16]; reducing cancer mortality, but not incidence, by over 30% in slim but not obese people overall, and reducing cancer incidence by 25% in deficient black Americans [from VITAL trial data giving 2000 IU/day] [17]; by reducing D2d risks as mentioned above; through reducing overall age-adjusted mortality by ~30% per 10 nmol/l genetically estimated increase in serum 25(OH)D in subjects whose 25(OH)D values were below 40 nmol/l [from Mendelian randomization analysis] [18];by reducing myocardial infarction rates by 27%; by reducing all- cause mortality [e.g. by 30% as seen in deficient Veteran Health Administration patients whose 25(OH)Ds were increased from <50 nmol/l to >75 nmol/l.] [19]

For Covid-19, prospective data from the large UK-Biobank cohort of middle-aged Britons showed a 34% reduction in infection rates in subjects reporting taking vitamin D supplements, but not other supplements at baseline, suggesting that improved vitamin D status could reduce Covid-19 infection rates prospectively [20]. Progressive reductions in Covid-19 infection rates were also reported with higher serum 25(OH)D values during the spring before the pandemic began in ~190,000 people across all 50 American states; those reductions plateauing out as 25(OH)D values approached 100 nmol/l at rates 50% lower than those seen with baseline 25(OH)Ds of 50 nmol/l, suggesting dose-effects of better vitamin D status rather than a specific threshold for this effect [21]. However, since those Covid-19 risk reductions were maximal for those with pre-pandemic 25(OH)Ds at or above 75 nmol/l [30 ng/ml], as are many other health benefits now emerging from further RCT data reanalyses [22] this level might provide a justifiable cutoff value for defining deficiency for non-bony tissues upon which updated public health guidance could be developed for vitamin D requirements. This might be at a minimum of 800 IU/day, or more, alongside a suitable national program for ensuring adequate adult vitamin D intakes at the population level [23].

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Funding

This paper was not funded.

Supplementary material

Supplemental data for this article can be accessed here