How Much Vitamin D Do You Actually Need? Sun, Skin, and the 20 ng/mL Line

Dhaka gets more than two thousand hours of sunshine a year, and the largest pooled estimate still puts two thirds of Bangladeshi adults below the deficiency line. A 2021 meta-analysis of South Asian studies found exactly that. "The highest prevalence of vitamin D deficiency was found in Pakistan (73%;95% CI: 63 to 83%) followed by Bangladesh (67%; 95% CI: 50 to 83%), India (67%; 95% CI: 61 to 73%), Nepal (57%; 95% CI: 53 to 60%) and Sri Lanka (48%; 95% CI: 41 to 55%), respectively." Plenty of sun, low numbers. Understanding why that happens is more useful than any single tablet recommendation, because it tells you which of the habits below will actually move your level and which ones will not.

Table of Contents

• 1. How Much You Need, and What the Labels Use
• 2. How Skin Makes It, and Why Sunny Countries Still Run Low
• 3. What the Blood Number Means
• 4. What Supplementation Actually Fixes
• 5. Infections, Diabetes and the Honest Numbers
• 6. How to Dose It Safely
• 7. Who Should Be Supplementing Without Guesswork
• 8. Food, Sun Habits, and Real Life
• 9. Frequently Asked Questions
• 10. The Short Version

How much you need, and what the labels use

The US Institute of Medicine framework, still the reference point most governments copy, sets requirements by life stage. The values are written to keep bones healthy and calcium metabolism normal, not to push blood levels to a trophy number. The supplement industry prefers higher targets, and that gap explains most of the confusion on this topic.

Group	Daily RDA	Tolerable upper limit
Infants 0 to 12 months	400 IU (10 mcg), as an adequate intake since breast milk alone does not cover it	1,000 IU
Children and teens 1 to 18 years	600 IU (15 mcg)	2,500 to 4,000 IU by age band
Adults 19 to 70 years	600 IU (15 mcg)	4,000 IU
Adults over 70	800 IU (20 mcg), because skin synthesis declines with age	4,000 IU
Pregnancy and breastfeeding	600 IU, the same as the non-pregnant adult number	4,000 IU

Food labels use a daily value of 20 micrograms, which is the 800 IU number, because the label was designed to cover the hardest group (older adults and pregnant people) rather than the average one. That single fact resolves most label anxiety. A product offering 400 IU is half the label value and still a meaningful dose for a younger adult. The upper limit sentence from the NIH fact sheet is worth quoting exactly. "The Tolerable Upper Intake Level for vitamin D ranges from 25 to 100 mcg (1,000–4,000 IU), depending on age." Note that the upper limit is not a target. It is the line above which routine daily use has no safety cushion.

How skin makes it, and why sunny countries still run low

Vitamin D is unusual among nutrients because the body manufactures most of what it needs, from ultraviolet light rather than food. Cholesterol in the outer skin layer absorbs UVB radiation and converts into the pre-vitamin, which the liver and kidneys then activate over the following two days. The system has a built-in ceiling. The same sunlight that makes the vitamin also breaks down the excess in the skin, which is why sun exposure cannot cause an overdose the way tablets can.

That mechanism is also why a sunny country can run a deficiency epidemic, and the list of reasons is mundane. UVB does not arrive equally through windows, clothing, shade, smog, or dark and melanin-rich skin, and a typical urban day in Dhaka, commute in an auto-rickshaw, office indoors, evening indoors, can easily produce less than the ten or fifteen minutes of midday arm-and-leg exposure the system needs. The NIH fact sheet states the baseline plainly, that "Most people in the world meet at least some of their vitamin D needs through exposure to sunlight". What that means for any individual adult depends on time of day, season, latitude, and skin pigmentation, and those four all work against the indoor urban routine. South Asian skin pigmentation is an efficient defense against sun damage and a real discount on vitamin D production, and the clothing and indoor-life habits stack on top of that.

What the blood number means

The test measures 25-hydroxyvitamin D, the storage form, and the debate about where to draw the line is not settled. The Institute of Medicine position, which the NIH fact sheet summarises, is that bone health is fine at ordinary levels: "Levels of 50 nmol/L (20 ng/mL) or more are sufficient for most people." Below that, "Some people are potentially at risk of inadequacy at 30 to 50 nmol/L (12–20 ng/mL)", and the outcome sentence is short. "Vitamin D sufficiency prevents rickets in children and osteomalacia in adults." The Endocrine Society publishes a higher cut-off of 30 ng/mL for "insufficiency", which is the number labs and clinics in Bangladesh usually print on reports, and it is also the number the supplement industry loves, because it converts more of the population into customers.

The practical reading of a report goes like this. A level of 21 ng/mL with no bone disease history is a number to raise modestly, not a medical crisis, and the gap between "insufficient" and "deficient" on that lab slip mostly reflects which society's cut-off the laboratory copied. A level under 10 ng/mL is a different situation and deserves a real treatment plan with follow-up blood work. Levels below 4 ng/mL are toxic territory in the other direction, which is the subject of the warning box later here.

What supplementation actually fixes

Start with what it reliably does, because the list is shorter than the marketing. Correcting true deficiency treats and prevents osteomalacia and rickets, helps calcium absorption, and over months can improve the non-specific muscle aches and weakness that marked deficiency causes. That is the core clinical use, and it has been known for a century, which makes it the least glamorous claim in the entire category.

The wider claims have aged less well. Large trials of vitamin D in generally healthy, not-severely-deficient populations have mostly been flat, including falls in community-dwelling older adults, blood pressure, and heart outcomes, and several hoped-for effects shrank once trials used adequate doses against placebo rather than comparing supplement users against sun-shy people. Fracture prevention shows benefit mainly when calcium is added in institutionalised older adults, and vitamin D alone in community walkers shows little. None of this argues against testing and treating deficient people, which remains clearly useful, but it does say the tablet is a correction for a shortfall, not a general performance tool.

Infections, diabetes and the honest numbers

The infection question got the best available answer in a 2017 individual-participant meta-analysis in the BMJ, which pooled 25 trials and eleven thousand people. The headline line reads "Vitamin D supplementation reduced the risk of acute respiratory tract infection among all participants (adjusted odds ratio 0.88, 95% confidence interval 0.81 to 0.96". Read carefully, that is a twelve percent reduction in respiratory infection risk across everyone supplemented, which is real and modest. The same analysis found the benefit was concentrated where the mechanism says it should be. People who started severely deficient improved far more, and the protection showed up with regular daily or weekly dosing rather than monthly mega-doses.

The diabetes question was answered in the other direction by the D2d trial, which gave 4,000 IU daily or placebo to people already diagnosed with prediabetes. Its conclusion. "Among persons at high risk for type 2 diabetes not selected for vitamin D insufficiency, vitamin D3 supplementation at a dose of 4000 IU per day did not result in a significantly lower risk of diabetes than placebo." The hazard ratio sat at 0.88 with confidence crossing unity, meaning a possible twelve percent benefit that the data could not confirm. A subgroup signal among the most deficient participants at baseline kept the trial interesting, and a new diabetes-prevention recommendation is not coming from a single trial with a null primary outcome.

Set the two results side by side and the logic is consistent. Supplementation helps outcomes that depend on correcting a deficit, and fades to noise when tested in people who were not short of it. Anyone selling the tablet as broad protection is quoting the deficient-persons numbers to replete people.

How to dose it safely

For adults wanting to move a 21 ng/mL number up without a treatment plan, the sensible range is 1,000 to 2,000 IU daily, taken with a meal containing fat because the vitamin needs dietary fat to absorb. That range sits well under the 4,000 IU upper limit, raises typical levels into the 30s within two to three months, and carries almost no monitoring burden. Higher daily doses, 4,000 to 5,000 IU, are used under supervision for documented deficiency, and the recheck belongs with them.

The weekly or monthly mega-dose schedules that pharmacies sell as convenience are the one dosing format the trials actively disfavour. The BMJ meta-analysis separated them and found the regular-dosing arms carried the benefit. A single 60,000 or 100,000 IU bolus produces a large, briefly supraphysiologic swing, and in elderly fallers one such annual mega-dose trial measured more falls and fractures in the bolus arm than in placebo. The prescription-strength sachets have their place, loading severe deficiency at the start of a supervised plan, and they are also the format people misuse by repeating them on Instagram's advice instead of a blood test.

Toxicity is rare but it is a real, specific syndrome, and the NIH fact sheet names it in one line. "Vitamin D toxicity is almost always a result of excessive intakes of vitamin D through supplements." The mechanism is calcium. Vitamin D drives calcium absorption, and too much of it drives hypercalcemia, with nausea, constipation, thirst, weakness, and in sustained cases kidney stones or worse. People with sarcoidosis, some lymphomas, or primary hyperparathyroidism can over-respond to ordinary doses because their disease makes its own active vitamin D, which is the exception that belongs in a doctor's hands rather than a spreadsheet.

Who should be supplementing without guessing

Some groups get a pass on the "test first" rule because the prior probability of shortfall is so high that trials and guidelines treat them together. Adults over sixty-five, because skin synthesis has fallen and the fracture stakes have risen. Pregnant and breastfeeding women, where deficiency costs are shared with the infant. People who cover up outdoors by clothing, work or convalesce indoors for weeks, or have darker skin at higher latitudes. Anyone with obesity, where the vitamin gets sequestered in fat tissue and circulating levels read low. Malabsorption states, coeliac and inflammatory bowel disease, post-bariatric surgery, chronic kidney disease, and long-term steroid or anti-seizure medications. Infants in the first year, breastfed, get 400 IU daily as standard in most country guidelines, and in Bangladesh that advice matters most for the dark-skinned, swaddled, indoor baby rather than the obviously deficient ones.

Everyone else with a vague fatigue complaint is guessing in both directions, and the cheap fix for that is the 25-OH test itself, which costs less than a year of the tablet in most cities and ends the argument.

Food, sun habits, and real life

Food sources of vitamin D are few and their amounts are fixed, which makes this the shortest food table in any health guide worth trusting.

Source	Typical vitamin D	Comment
Salmon, cooked, 100 g	350 to 550 IU	Farmed salmon sits at the lower end, wild sockeye at the higher, and it is the only commonly eaten food that comes close to a dose in one serving.
Canned sardines or tuna, 100 g	150 to 250 IU	The realistic daily option where fresh salmon is absent or priced like meat, and the oil-packed version absorbs the vitamin with the fat already handled.
Egg yolk, one large	35 to 45 IU	Free-range and sun-exposed hens produce several times more, which is the one legitimate egg-label claim in this category.
Fortified milk or oil, 1 cup	100 to 120 IU	Fortification policy varies by country. Read the carton, and note that most South Asian milk is not fortified at all.
Mushrooms grown in sunlight, 100 g	Variable, up to 400 IU	The only plant source with meaningful vitamin D, and only when the grower exposed them to UV. Ordinary dark-grown mushrooms have none.

On the sun side of the ledger, the practical habit that the South Asian data supports is frequency rather than duration. Ten to fifteen minutes of midday sun on forearms and lower legs, most days, produces enough precursor for the average adult without a burn, and the melanin discount discussed above is exactly why that number is higher than what fair-skinned populations need. Morning and evening sun, however pleasant, makes almost no vitamin D because the UVB fraction that does the job exists only when the sun is high. A car window or office window filters most of that band out, so the commute through glass builds nothing.

Frequently asked questions

Should everyone take a vitamin D supplement just in case?

Not everyone needs a tablet, but at these latitudes with these indoor habits the gap between default intake and requirement is small and the safe range is wide, so 1,000 IU daily without testing is a defensible habit for an otherwise healthy adult. The reason testing still comes first in this guide is that a 21 ng/mL level and a 7 ng/mL level get genuinely different plans, and the tablet alone cannot tell you which person you are.

Is the 60,000 IU weekly sachet bad?

As a short supervised loading phase for severe deficiency it is a standard, effective tool, and the evidence against it concerns maintenance use. A trial in older women that gave a large yearly bolus found more falls, which is how the weekly mega-dose habit lost its standing. If the plan comes from a doctor with a date to stop, follow it. If it comes from a pharmacy habit repeated forever, that is the pattern the data dislikes.

Do tanning beds or morning sun fix it?

No on both, for the same physics. Tanning beds emit mostly UVA, which browns skin without producing the UVB reaction that makes vitamin D, and morning and evening sunlight arrives through too much atmosphere for the UVB band to survive the trip. The window that works is roughly when your shadow is shorter than you are.

The short version

The requirement is 600 IU a day for most adults and 800 past seventy, and the number printed on the label, 20 micrograms, is the ceiling version for the most vulnerable groups. Two thirds of adults in this region sit under the deficiency line despite the sunshine because clothing, shade and indoor schedules beat latitude. A blood test settles the question for the price of a few tablet packs, and the honest reading of a 21 ng/mL result is raise it gently, not panic. Supplementation reliably treats true deficiency and cuts infection rates in deficient people, does not much for the already-sufficient, and 1,000 to 2,000 IU daily with a fatty meal is the safe middle of the road for anyone supplementing without a plan. The mega-dose sachets belong to treatment, not maintenance, and the ten-minute midday habit is the cheapest supplement in the market.

Read next

• How much iron you actually need, ages, sex, and what ferritin means
• The best time for your flu shot and why October beats July
• How much sleep per night, what your age bracket really changes

Sources

1. National Institutes of Health, Office of Dietary Supplements. Vitamin D: Fact Sheet for Health Professionals. ods.od.nih.gov
2. World Health Organization and national guidelines on infant vitamin D, as summarised by the NIH fact sheet reference set. who.int
3. Martineau AR, et al. "Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data." BMJ 2017. PMID 28202713.
4. Pittas AG, et al. "Vitamin D Supplementation and Prevention of Type 2 Diabetes." N Engl J Med 2019. PMID 31173679.
5. Siddiqee MH, et al. "High prevalence of vitamin D deficiency among the South Asian adults: a systematic review and meta-analysis." BMC Public Health 2021. PMID 34627207.