Vitamin D Deficiency and Its Health Consequences in Africa.

Africa is heterogeneous in latitude, geography, climate, food availability, religious and cultural practices, and skin pigmentation. It is expected, therefore, that prevalence of vitamin D deficiency varies widely, in line with influences on skin exposure to UVB sunshine. Furthermore, low calcium intakes and heavy burden of infectious disease common in many countries may increase vitamin D utilization and turnover. Studies of plasma 25OHD concentration indicate a spectrum from clinical deficiency to values at the high end of the physiological range; however, data are limited. Representative studies of status in different countries, using comparable analytical techniques, and of relationships between vitamin D status and risk of infectious and chronic diseases relevant to the African context are needed. Public health measures to secure vitamin D adequacy cannot encompass the whole continent and need to be developed locally.

Introduction

This article addresses the prevalence of vitamin D deficiency and issues that need to be considered when considering vitamin D status (supply, sufficiency, utilisation) and potential health consequences in Africa. The reader is referred to other articles that discuss in detail some of the general background to vitamin D, including evolution, biologic functions and dietary requirements; photobiology; assays; pregnancy and lactation; calcium and vitamin D deficient rickets and children’s bone health; dietary sources; sun deprivation and health consequences. Where appropriate these are also considered here because they are particularly relevant to, or may be different in, an African context.

Vitamin D status and supply is influenced by sunshine exposure, diet and underlying health conditions. People acquire vitamin D by cutaneous synthesis following exposure to sunlight, and from the diet. It is not possible to generalize about the proportions contributed by different sources because this varies considerably around the world, across population groups, and between individuals, mainly because of wide differences in skin exposure to ultraviolet B (UVB) radiation, the efficiency of cutaneous synthesis, dietary supplementation and food fortification practices [1]. Reliance on dietary sources of vitamin D is greatest during the winter at latitudes outside the tropics (>30°N and >30°S) and among people with restricted skin sunshine exposure [1, 2]. The endogenous supply of vitamin D3 (cholecalciferol) depends on the skin’s exposure to UVB radiation at wavelengths of 290–315 nm, the efficiency of cutaneous vitamin D synthesis, and the extent to which vitamin D is degraded within the skin [3-5]. The quantity of UVB radiation of the relevant wavelengths that reaches the skin depends on many factors (see section “Ultraviolet radiation and skin pigmentation”) including time of day, and, at latitudes north and south from approximately 30°N and 30°S, on month of the year. In the summer months, however, the amount of UVB containing wavelengths that can generate vitamin D synthesis is the same at all latitudes [6]. In addition, even when there is abundant sunshine, the extent of UVB skin exposure depends on clothing, living and working environments and sunscreen use [7].

The status marker for vitamin D, plasma 25 hydroxy vitamin D (25OHD) concentration, depends on factors that determine the balance between supply and turnover [8]. Supply depends on factors including host, dietary and environmental factors. Turnover of 25OHD and 1,25 dihydroxy vitamin D (1,25(OH)2D) in renal and extra-renal tissues depends on many factors, including nutrient intake (of which calcium is the most important), protein-energy malnutrition, body composition, high disease burden and innate characteristics such as polymorphisms in the VDR and DBP genes [8-12].

The Continent of Africa

Geography, Topography, and Climate

Africa is the world’s second largest continent after Asia, occupying an area of about 30 million km2 including adjacent islands. Thus, Africa represents about 20% of the earth’s total land area. It is the only continent that straddles the equator and has both northern and southern temperate zones. The distance between the most northerly (~37°N in Tunisia) and most southerly (~34°S in South Africa) points is ~8000 km. From the westernmost (~17°W, Cape Verde) to the easternmost (~51°E in Somalia) points is ~7400 km. Some of the larger African countries are distributed across more than one latitude band. The climate ranges from tropical to subarctic and the topography includes deserts, mountains, grassland plateaus, lakes and rivers. Much of northern Africa is predominantly desert or arid land, and central and southern areas contain deserts, savannah plains and dense rainforest. All these features affect the available water supply, food sources, population distribution and the sorts of clothing worn (e.g. for warmth, protection against the sun, wind, rain and sand). The weather in Africa is also very variable, and temperatures vary greatly throughout the continent. Many countries have wet and dry seasons which in turn affect food availability, farming activities and therefore levels of physical activity and weight, and body composition changes, morbidity and mortality, including incidence of low birthweight infants, and the prevalence and incidence of malaria. About 74% of people in Africa live in areas that are highly endemic for malaria, 19% in epidemic-prone areas and only 7% in low-risk or malaria-free areas [13].

Demography

Africa is the second most populated continent after Asia with more than 900 million people who account for ~14% of the world’s population. Africa comprises more than 50 countries ranging in landmass from 2.4 million km2 (Sudan) to 455 km2 (The Seychelles) (www.fao.org). Individual counties are not only very diverse with respect to size but also population ranging from 145,000,000 (Nigeria) to 86,000 (The Seychelles) and the proportion of the population that live in urban areas (10% Burundi; 84% Gabon; average 38% in WHO Africa Region) [14]. Across the continent as a whole ~40% of Africans are Christian, ~40% are Muslim and ~20% follow other religions (including indigenous African religions, Hinduism and Judaism). There is a great deal of variation within and between countries and population groups. Some religious beliefs and traditional practices can impact on customary dress and dietary habits.

Ultraviolet Radiation and Skin Pigmentation

The factors that affect the amount of ambient UVB radiation include stratospheric ozone levels, cloud cover, latitude, season, and lower atmospheric pollution. The factors that affect an individual’s UVB exposure and biological effects of UVB include sun-seeking and sun protective behaviours, skin pigmentation, and cultural dress and behaviour [15]. The concentration of atmospheric ozone that is able to absorb UVB, and the amount and spectral structure of radiation reaching the body is dependent on the angle at which the sun’s rays pass through the atmosphere; at lower latitudes, closer to the equator there is more intense radiation. At higher altitudes, UVB radiation is increased because there is less mass of air for solar radiation to pass through. Time of day, season and presence of clouds, dust, haze and various organic compounds can alter the intensity of incident solar radiation. The UV Index, the maximum UV level on a given day during the 4-h period around solar noon, varies according to time and place, and with season and latitude. Levels of total annual UV radiation (UVR) vary approximately fourfold globally, but within any area there can be a tenfold difference in personal exposure related to behavioural and cultural factors. Individual exposure within population groups may vary from one tenth to ten times the mean exposure in a particular location.

Different skin pigmentations are thought to have developed to protect against high ambient UVR whereby those inhabiting low latitudes with high UVR have darker pigmentation to protect against deleterious effects, whilst those inhabiting higher latitudes developed fair skin to maximize vitamin D production from much lower ambient UVR. However, in recent history, migration of people of different skin types means that skin pigmentation is not necessarily so suited to the environment. For the purposes of assessing the effects of UVR on different skin pigmentations, a scale classifying skin types for UVR sensitivity is used which comprises six skin types [16, 17]. This is often broken down into three broader skin pigmentation groups [15]: lightly pigmented (Caucasian fair, medium and darker skin, skin types I–IV), intermediate pigmentation (Asian or Indian skin, skin type V), and deeply pigmented (Afro-Caribbean or black skin, skin type VI). Latitude provides a rough approximation to global variation in UVR, so in epidemiological studies of UVR, groups of different races/ethnicities can be categorized under these broad headings, but with recognition that there is great individual variation within racial and ethnic groups [15]. An illustration of the distribution of skin types at different latitudes in Africa is shown in Table 1.

Table 1

WHO sub-regionsa comprising African countries, by latitude, and % distribution of population within a region in that latitude with different skin pigmentation (compiled from information in Ref. [15])

Latitude	WHO sub-region, country (% of country in latitude band, if not 100%)	Lightly pigmented	Intermediate pigmentation	Deeply pigmented
0–10	Africa D: Cameroon, Equatorial Guinea, Gabon, Ghana, Liberia, Sao Tome and Principe, Seychelles, Sierra Leone, Angola (50%), Benin (90%), Nigeria (50%), Guinea (50%), Togo	<1	<1	41
	Africa E: Burundi, Central African Republic, Congo, Cote D’Ivoire, Kenya, Rwanda, Democratic Republic of Congo (95%), Ethiopia (50%), Tanzania	≪1	<1	59
	East Mediterranean D: Somalia, Sudan (20%)	<1	<1	3
10–20	Africa D: Cap Verde, Chad, Comoros, Gambia, Guinea-Bissau, Mauritius, Niger, Senegal, Angola (50%), Benin (10%), Burkina Faso, Guinea (50%), Madagascar (50%), Mauritania (50%), Nigeria (50%), Mali	<1	2.5	42
	Africa E: Eritrea, Malawi, Zimbabwe, Botswana (10%), Democratic Republic of Congo (5%), Mozambique (70%), Namibia (20%), Ethiopia (50%)	<1	<1	22
	East Mediterranean D: Sudan (79%)	8	1	4
20–30	Africa D: Algeria (10%), Madagascar (50%), Mauritania (50%)	<1	2	2
	Africa E: Lesotho, Swaziland, Zambia, Botswana (90%), Mozambique (30%), Namibia (80%), South Africa (80%)	1	1	13
	East Mediterranean B: Libya (50%), Morocco (10%)	22	2	<1
	East Mediterranean D: Sudan (1%), Egypt (50%)	<1	34	<1
30–40	Africa D: Algeria (90%)	<1	10	<1
	Africa E: South Africa (20%)	<1	<1	2
	East Mediterranean B: Libya (50%), Morocco (90%), Tunisia	70	2	<1
	East Mediterranean D: Egypt (50%)	14	34	<1

aCountries are listed according to both WHO Africa regions (D and E) and other countries on the African continents which come under WHO East Mediterranean regions B and D. All these WHO regions encompass the countries mentioned in this article

The efficiency of vitamin D cutaneous synthesis depends on skin pigmentation and age [18, 19]. Very little data from controlled interventional research are available on the differences in efficiency in vitamin D synthesis between different skin types, but a fourfold difference has been reported between types II and V [20]. However, evidence from studies in African-American Blacks and Indians suggest that there is no difference in the total capacity of cutaneous vitamin D synthesis between individuals with different skin types [21, 22]. There are no studies in the literature investigating cutaneous synthesis in indigenous African people.

On the African continent, seasonality in cutaneous vitamin D synthesis would be expected in countries such as Morocco, Tunisia, Algeria, Libya, Egypt, and South Africa which lie at latitudes >30°N and >30°S [23]. This is supported by studies in hip fracture patients in South Africa [24] and in pregnant women and their newborns in Algeria [25]. In countries close to the equator, seasonality in cutaneous vitamin D synthesis might be expected due to differences in cloud cover (http://nadir.nilu.no/,olaeng/fastrt/VitaminD.html). In two studies, in The Gambia (14°N) and Guinea Bissau (12°N), no significant effect of season on plasma 25OHD concentrations was found [26, 27]. However, seasonal variations in plasma 25OHD concentration have been reported in older people in The Gambia where lower concentrations were observed in December–February [1].

Diet and Foodstuffs

The African diet is very heterogeneous and varies from region to region. Historically, the climate, geography and history largely dictated where populations became pastoralists, agriculturalists and hunter-gatherers, and therefore local availability and ability to store fresh and salt-water fish, meat, poultry, and milk; and the kinds of crops that could be grown and thus traditionally consumed. Over the centuries there have been many secular changes in diet and food composition. There are now also ‘westernised’ populations in most African countries; the proportions differ between and within countries and are changing as countries transit from less to more developed, and depending on natural and manmade environment. This transition affects dietary patterns, the sorts of foods consumed, nutrient intakes and sources of nutrients.

Indicators of Health and Burden of Disease in Africa Which May Affect, or be Affected by Vitamin D Status

Africa is associated with a high prevalence of economic disadvantage and high burden of disease. This is still mostly infectious disease but increasingly also non-communicable disease. The various factors that may be especially important to consider in the context of Africa and that may affect the interpretation of vitamin D status and deficiency are outlined later.

In addition, vitamin D status has been implicated in the progression of disease, and this can add another layer of complexity in interpreting vitamin D status measurements. The emerging work on the possible relationship between vitamin D status and disease risk reduction (TB and HIV/AIDS) is discussed in section “Health consequences of poor vitamin D status In Africa”.

Malnutrition

The percentage of infants born with low birthweight is ~14% in the WHO African region and in some African countries is >20%. There are many reasons for low birthweight but a major contributory factor in Africa is low maternal weight gain due to general shortage of foods or energy or specific nutrients. Low birthweight is the major risk factor for infant mortality and morbidity. The most recent estimate of infant mortality rate in the WHO Africa region is 94 per 1000 live births. In some African countries the rate is >100. These figures are in the context of a global rate of 49. The under 5 mortality rate in Africa is 157 per 1000 live births and in some African countries is >200 per 1000 live births (global rate = 71). Recent estimates suggest that 43% of children in Africa are stunted and 23% are underweight for their age. In some African countries, the prevalence is >50% and >30%, respectively [14]. Vitamin D or calcium deficiency rickets are often found where more general malnutrition is prevalent.

Vitamin D Intake

There are very little data from Africa on dietary vitamin D intakes, although sources of other nutrients and food patterns indicate that, as observed elsewhere, food intake is unlikely to make a substantial contribution because there are few naturally occurring food sources that are rich in vitamin D (meat, offal, egg yolk, oily fish), and these are not eaten at all or only rarely [28-30]. It is impossible to quantify or generalise about the possible contribution of vitamin D in foods supplied as food aid (some of which may be fortified) or in the form of multiple micronutrient supplements.

Low Calcium Intake

Where dietary calcium intakes have been assessed in African countries, the values from infants, children and adults are considerably below recommendations, and close to biological requirements. Calcium intakes of 200–300 mg/day have been reported from Egypt, Kenya, Nigeria, The Gambia and South Africa [31]. Dairy product intake is minimal in many African counties and typical diets also contain high amounts of phytates, oxalates and tannins that are likely to reduce the absorption of calcium. Children with bone deformities consistent with rickets but without overt vitamin D deficiency have been reported from Nigeria, The Gambia and South Africa. Dietary calcium deficiency has been identified as a possible contributing factor due to its effects on the metabolism of vitamin D [32].

Fluorosis

Fluoride is present in all natural waters, but higher concentrations are found in groundwater from calcium-poor aquifers and in areas where fluoride-bearing minerals are common. In Africa the latter applies particularly to areas in the East African rift system and many lakes in the Rift Valley (Sudan, Ethiopia, Uganda, Kenya and Tanzania) have very high levels of fluoride. High groundwater fluoride concentrations are associated with igneous and metamorphic rocks including those in western Africa (e.g. Nigeria, Senegal) and South Africa [33]. Endemic skeletal fluorosis has been associated with hypocalcaemia, phosphaturia and raised concentration of 1,25(OH)2D [34]; all indices that are associated with various forms of rickets.

Tropical Enteropathy

Tropical enteropathy (tropical sprue) is common in many parts of Africa. It is characterised by intestinal permeability and poor absorption, abnormal villus structure and function in adults and children and thought to be caused by many factors including repeated episodes of infection and diarrhea [35-39]. Poor nutrient absorption of calcium and other factors that may affect vitamin D metabolism may impact on vitamin D status, which in turn may affect gastrointestinal health. Thus tropical enteropathy may be a contributing factor in vitamin D deficiency and conversely may be exacerbated by poor vitamin D status. However, although these links are plausible, they have not yet been explored.

Malaria

Of the 350–500 million estimated cases of clinical malaria each year, nearly 60% of cases and 90% of deaths occur in Africa south of the Sahara with 75% of these in children under 5 [13]. Malaria affects everyone, but children under 5 years of age and pregnant women are the most vulnerable groups because of lower immunity. Malaria contributes to health problems and deaths in many ways, especially in younger children. These include frequent acute infections; anaemia as a result of repeated or chronic malaria infection; malaria in pregnancy resulting in low birth weight in infants; increased susceptibility to other diseases such as respiratory infections and diarrhea [40]. The immunomodulatory properties of vitamin D discussed in section “Health consequences of poor vitamin D status In Africa” may also play a role in the progression of malaria and its co-morbidities.

HIV/AIDS

Sub-Saharan Africa is the region most affected by HIV/AIDS. In 2007 one in three people in the world with HIV lived in sub-Saharan Africa; a total of 22.5 million people. Cause-specific mortality rates per 100,000 population in 2006 were 203 for HIV/AIDS (compared with a global rate of 34); 56 for TB among HIV-negative people (global rate 22); 26 for TB among HIV-positive people (global rate 4). The mean prevalence in Africa (estimated in 2005) of HIV among people over 15 years of age is 4459 per 100,000 population (global figure is 662) and in some African countries the figures are >20,000 [14].

Tuberculosis

Globally tuberculosis (TB) is the second highest cause of death from infectious disease and is responsible for more deaths in those individuals infected with HIV than any other infectious disease [41]. Almost 30% of TB cases and 34% of TB-related deaths occur in Africa, and it is estimated that between 1990 and 2005 the incidence of TB more than doubled in many African countries [42]. The prevalence of TB in Africa (estimated in 2006) is 547 per 100,000 population, and is >900 in some African countries (global prevalence is 219) [14]. Although HIV infection is a major cause, poverty, inadequate healthcare (in terms of diagnosis and treatment) and drug resistance have all contributed [42].

Non-Communicable Disease

In addition to a high burden of infectious disease in Africa, secular changes in diet and lifestyle, particularly in North Africa and urban centres in sub-Saharan Africa, are leading to increases in cardiovascular disease, cancer, and obesity and its co-morbidities including hypertension and type 2 diabetes [43-47]. Low vitamin D status has been implicated in diabetes and cancer. Serum 25OHD has been found to be low in obese adults and may be due to sequestration of vitamin D in subcutaneous fat and its consequent reduced bioavailability [48]. Currently, cancer is estimated to cause only 4% of deaths in Africa, but because of large populations this equates to a large number of people, and furthermore, many cases are undiagnosed. The major cancers in African men are Kaposi’s sarcoma (associated with HIV), liver, prostate, bladder, and oesophageal cancer, and non-Hodgkins lymphoma. In African women the major cancers are Kaposi’s sarcoma, cervical, breast, liver, and ovarian cancer, and non-Hodgkins lymphoma [49].

Health Consequences of Poor Vitamin D Status in Africa

The classical effects of poor vitamin D status are rickets and osteomalacia, but there is also emerging evidence for a role of vitamin D in reduction in the progression or severity of TB and HIV/AIDS. The immunomodulatory effects of vitamin D means there are many potential health consequences of vitamin D deficiency in Africa where the infectious disease burden is high. Vitamin D deficiency may impact on the immune system by decreasing innate immunity and immune surveillance; decreasing T lymphocyte number and function; and disrupting the Th1/Th2 balance (vitamin D normally inhibits Th1 profile). Vitamin D deficiency in an African setting may impact on the progression of infectious diseases such as TB (see below), HIV (see below) and schistosomiasis [50].

Rickets and Osteomalacia

Well-established health consequences of vitamin D deficiency are rickets and osteomalacia. However, although vitamin D deficiency rickets is reported in Africa, there are very little data on the burden of disease this imposes, and on associations with malnutrition, poverty and other factors. It is therefore difficult to judge the extent to which primary vitamin D deficiency is a cause.

In Europe, as early as 1849, cod liver oil was recognised as beneficial in the treatment of TB and led in the early 1900s to the widespread use of vitamin D as a therapeutic treatment for both cutaneous and pulmonary TB [51]. The use of sanatoria was based on the belief that fresh air and sun exposure led to a positive outcome in the treatment of TB. Recently, a potential mechanism has been elucidated. In an in vitro study using serum from African-Americans and Caucasians, it was demonstrated that recognition of mycobacterium tuberculosis particles by toll-like receptors on the surface of macrophages triggers the upregulation of VDR and CYP27B1 (1 alpha hydroyxlase) expression. Intracellular 1,25(OH)2D production and signalling upregulated production of the antimicrobial peptide, cathelicidin, that is able to destroy the intracellular mycobacterium. Importantly, the serum from African-Americans (~20 nmol/l 25OHD vs. ~80 nmol/l in Caucasians) was less efficient in upregulating cathelicidin [52].

Observational studies have demonstrated a link between low serum 25OHD concentration and susceptibility to TB [53-58], and patients with active TB have lower serum 25OHD concentration than contacts (tuberculin skin-test positive) from similar ethnic and social backgrounds [59]. Despite the high incidence of TB in Africa, only one study, from Guinea-Bissau (latitude 12°N), has been published that compares patients with active TB and a control group. The latter consisted of both tuberculin skin test positive and negative subjects [27]. There was a high prevalence of plasma 25OHD <75 nmol/l in both groups but this was more common in patients than controls, although not after correction for socio-economic and demographic factors. However, mean serum 25OHD concentration remained lower in the TB group. A systematic review of studies conducted around the world, including in Kenyans, has recently concluded that low serum 25OHD is associated with greater risk of active TB [60]. However, although evidence suggests that lower vitamin D status is associated with increased risk of TB, none of the studies could disentangle cause and effect because of relatively small sample size.

Studies have investigated polymorphisms in the vitamin D receptor (VDR) gene in relation to TB and other diseases in Africa and elsewhere. The polymorphisms Taq1 and Fok1 affect transcription and expression of the VDR protein and may result in functional effects, including susceptibility to TB, response to anti-TB treatment, and potentially vitamin D metabolism. Furthermore, because the VDR interacts with vitamin D metabolism, the effect of polymorphisms may depend on vitamin D status. Studies in West Africa [61-63], South Africa [64, 65] and East Africa [66, 67] have investigated VDR polymorphisms and TB. However, results were inconsistent, and a systematic review and meta-analysis of VDR polymorphisms and TB concluded that results were inconclusive, studies included heterogeneous populations and were underpowered [68].

A further area of potential study is the use of vitamin D as an adjuvant to TB drug treatment [69]. A study in Egypt [70] in which vitamin D was administered to children with TB concluded that clinical outcome in patients that received vitamin D plus drug treatment was better than those who received treatment alone, although differences were not significant. A double-blind, placebo-controlled vitamin D supplementation trial was performed in Indonesia in patients with moderately advanced pulmonary TB. Those receiving vitamin D had a significantly higher rate of sputum conversion and increased radiological improvement compared with the placebo group [71]. In vitro work has shown a positive effect of vitamin D; a single dose of 2.5 mg vitamin D was found to protect against reactivation of latent TB infection in whole blood from TB contacts [72]. A recent review [51] identified 14 reports of vitamin D administration to patients with pulmonary TB and concluded that major shortcomings in the studies prevented any clear conclusion from being drawn. Despite the strong mechanistic evidence for a possible beneficial role of vitamin D in the prevention and cure of TB, appropriately powered, randomised controlled trials are needed, especially in Africa where the disease burden is high. In the next few years there should be more definitive evidence because a number of registered clinical trials (http://www.controlled-trials.com) are investigating the effects of adjuvant vitamin D therapy on recovery from TB. One of these is a randomised, placebo-controlled double-blind trial [ISRCTN352121232] to investigate the therapeutic effects of oral vitamin D3 in 500 TB patients in Guinea-Bissau.

The role of vitamin D in the progression of HIV infection is largely unknown and studies have mostly been performed in Western settings. A recent review considers in detail the evidence for a role of vitamin D in HIV infection [73]. Lower concentrations of 25(OH)D and 1,25(OH)2D among HIV-infected people have been attributed to defects in renal hydroxylation and increased utilization. An inverse association between 1,25(OH)2D concentrations and mortality has been reported from a small cohort of HIV-infected adults, and some cross-sectional studies have indicated positive correlations between 1,25(OH)2D and CD4+ cell counts.

Poor vitamin D status may compound the negative effects of HIV infection and HIV anti-retrovirals on the skeleton (as evidenced by high incidence of osteoporosis and osteopenia in infected individuals). Drugs used in HIV treatment may inhibit or stimulate cytochrome P450 enzymes. Vitamin D and its metabolites are catabolised by P450 enzymes and so stimulation may lead to vitamin D deficiency [62]. More studies are needed to confirm the associations between vitamin D status and HIV disease progression.

Studies of Vitamin D Status in Africa

Plasma 25OHD concentrations measured in studies conducted in Africa are summarised in Table 2. The data are from studies in healthy men and women, elderly people, pregnant and lactating women, healthy children and those with rickets, and clinical studies of TB and pneumonia. The subjects encompass people of different religions and in different settings. Full details of the studies can be found in the original references.

Table 2

Studies that include assessment of vitamin D status conducted in Africa

Location and reference	Study group	Age	Plasma 25OHD (nmol/l) mean ± SD (if given), or median and range (n = number of subjects where specified)
North Africa
Algeria [25]	Healthy women at term delivery:	Not specified
	Summer		11.8 ± 4.2a (n = 56)
	Winter		9.0 ± 2.8 (n = 28)
	Newborns:	5 days
	Summer		9.0 ± 5.6 (n = 56)
	Winter		7.6 ± 2.1 (n = 28)
Egypt [77]	Healthy children	0–3 y	25.3 ± 10.3 (n = 14)b,r
	Active rickets		9.3 ± 7.3 (n = 31)
Ethiopia [78]	Healthy men and non-pregnant women (24 M, 6 F)	20–22 y	23.5 (range 18–29)c
	Healthy women at term delivery (n = 31)	22–28 y	25 (range 17–46)
Libya [79]	Children with active rickets (10 M, 6 F)	0.25–2 y	24.9 ± 22.4d,r
Libya [80]	Healthy women at term delivery	26–45 y	34 (13–75, n = 19)d,r
	Cord blood		20 (10–45, n = 14)
Tunisia [81]	Healthy veiled or non-veiled non-pregnant women	20–60 y	Veiled: 35.7e
			Non-veiled: 42.5
Tropical Africa
Gabon [82]	Healthy term infants (44 M, 35 F) at 0, 3, 6 months	0–0.5 y	0 mo: 109.96 ± 42.5a
			3 mo: 148.0 ± 54.3 (n = 27)
			6 mo: 150.5 ± 64 (n = 28)
The Gambia [1] from data collected by Dibba et al. [32, 83]	Healthy children	8–12 y	95.0 ± 19.6 (n = 44)f,g
	Children with non-active rickets	1–14 y	50.7 ± 12.8 (n = 30)
	Children with active rickets		42.4 ± 13.8 (n = 13)
The Gambia [26]	Lactating women 3 months postpartum	16–41 y	All year: 64.9 ± 18.5 (n = 28)e
			Jan–Mar: 64.1 ± 14.1 (n = 9)
			Apr–Jun: 66.1 ± 21.6 (n = 9)
			Jul–Sep: 54.7 ± 15.0 (n = 5)
			Oct–Dec: 73.9 ± 21.7 (n = 5)
The Gambia [84]	Healthy women	25–44 y	80.9 ± 22.8 (n = 11)e
		45–49 y	113.3 ± 26.5 (n = 12)
		50–54 y	95.7 ± 2.3 (n = 14)
		55–59 y	83.6 ± 19.4 (n = 21)
		60–64 y	97.7 ± 26.5 (n = 27)
		65–69 y	87.3 ± 25.1 (n = 13)
		70–74 y	87.0 ± 32.7 (n = 7)
		75+ y	72.3 ± 19.8 (n = 8)
The Gambia [85]	Healthy men and women	60–75 y	M: 64.3 ± 15.5 (n = 15)f,g
			F: 72.8 ± 17.5 (n = 15)
Guinea Bissau [27]	Healthy men and women (239 M, 255 F)	Not specified
	TB patients (221 M, 142 F)		85.3 ± 34.8g,h
			78.3 ± 22.6
Kenya [55, 60]		Mean (SD)	Median and range
	Healthy controls (8 M, 7 F)	33 ± 7.4 y	65.5 (26.25–114.75, n = 15)i,r
	TB patients (9 M, 6 F)	35 ± 8.3 y	39.75 (6.75–89.25, n = 15)
Nigeria [86, 87]	Healthy age matched children (9 M, 3 F)	3–5 y	41 (29–50)j,r
	Active rickets (7 M, 3 F)		36 (22–84)
Nigeria [88]	Healthy children	1–5 y	69 ± 22 (n = 20)e
	Children with active rickets		36 ± 28 (n = 22)
Nigeria [89, 90]	Healthy children	1–5 y	63 ± 17.8 (n = 47)e
	Children with active rickets		43 ± 33.5 (n = 37)
Nigeria [91]	Healthy children (matched for sex, age, religion, 19 M, 8 F)	0.8–7 y	60.0 ± 18.8 (n = 27)f
	Active rickets (12 M, 4 F)		35.3 ± 11.8 (n = 16)
Nigeria [92–94]	Healthy children	Median 42 mo	51.3 ± 15.5e
	Active rickets	Median 46 mo	34.8 ± 25.5
Nigeria [95, 96]	Healthy children (6 M, 4 F)	1–8 y	52.3 ± 7.3e (n = 10)
	Active rickets (3 M, 7 F)		24 ± 11.3 (n = 10)
Nigeria [97–99]	Healthy children (6 M, 9 F)	2–8 y	37.5 ± 11.5k (n = 15)
	Active rickets (6 M, 9 F)		37.5 ± 13.5 (n = 15)
Nigeria [100]	Healthy children (5 M, 4 F)	0.5–5 y	130 ± 107 (n = 9)
	Children with pneumonia (15 M, 9 F)		104 ± 59 (n = 18)
Nigeria [101]	Children’s survey (97 M, 121 F)	6–35 mo	All: 64.3 ± 23.3e (n = 218)
			Healthy: 65 ± 24 (n = 198)
			Rickets: 56.5 ± 11.8 (n = 20)
Nigeria [102]	Active rickets (9 M, 7 F)	1.2–2.0 y	28.5 (range 17–40)f
Nigeria [87, 103]	Healthy women at term delivery (10 in purdah, 20 not in purdah)	Not given	Mean and range
			Purdah: 53 (37–64)l,r
			Non-purdah: 90 (68–150)
	Cord bloods		Purdah: 31 (24–59)
			Non-purdah: 58 (35–79)
Zaire [104]	Healthy men (n = 33)	Mean 31 y	65 ± 39m,r
Southern Africa
South Africa [105]	Children with active rickets (2 M, 6 F)	4–13 y	48.9 ± 10.1 (n = 8)n,r
South Africa [24]	232 patients with hip fracture	72.7 ± 13 y	All: 44.3 ± 23n,r
			Nov–June (s): 51 ± 26.9
			June–Oct (w): 38.1 ± 17.0
South Africa [106]	Survey of school children	7–12 y
	M/F rural		72.3 ± 21.8 (n = 60)n,r
	M/F suburb		77.3 ± 23 (n = 60)
	M/F urban		82.8 ± 18.8 (n = 60)
South Africa [107]	Active rickets (2 M, 2 F)	4–14 y	62.5 ± 21.5 (n = 4)n,r
South Africa [108]	Survey of pre-school children	3–5 y	85.5 ± 19 (n = 20)o,r
South Africa [109]	Healthy black children	6–18 y	6–9 y: 123.9 ± 12.0 (n = 17)e
			10-13 y: 115.6 ± 5.7 (n = 26)
			14-18 y: 90.2 ± 10.8 (n = 15)
	Healthy albino children	6–18 y	6-9 y: 102.9 ± 11.4 (n = 30)
			10-13 y: 86.1 ± 14.0 (n = 36)
			14-18 y: 90.7 ± 3.4 (n = 16)
South Africa [110]	Children:	1–12 y
	Active vitamin D or Ca deficiency rickets		32.3 ± 18.8 (n = 21)o,r
	Hypophosphataemic rickets		52.5 ± 12.3 (n = 9)
	Healing rickets		61.8 ± 6.0 (n = 3)
South Africa [111]	Active rickets:	Median (range)
	Calcipenia	9.5 y (1.7–18.0)	38.5 (15.3–111.5)p,r
	Phosphopenia	5.7 y (0.3–16.0)	26.0 (10.0–62.5)
South Africa [112]	Healthy women at term delivery	16–40 y	81.75 ± 28.8 (n = 43)q,r
	Cord blood		170.97 ± 72.6 (n = 43)
South Africa [113]	Elderly female nursing home residents	80 ± 4 y	32 ± 11 (n = 60)a
South Africa [114]	Survey of women:	20–64 y
	Premenopausal black		48.3 (17.0–114.0, n = 74)n,r
	Premenopausal white		65.8 (34.0–114.8, n = 105)
	Postmenopausal black		47.5 (15.8–80.8, n = 65)
	Postmenopausal white		64.5 (25.5–139.8, n = 50)

y years, mo months, M male, F female

aNot detailed

bCompetitive binding assay [115]

cHPLC

dCompetitive binding assay after extraction and HPLC

eRIA Incstar

fRIA Diasorin

gParticipation in QA scheme DEQAS reported

hLC-MS/MS

iRIA [116]

jRIA after extraction [117]

kRIA Nicols after ethanol extraction

lCompetitive binding assay [118]

mCompetitive binding assay [119]

nCompetitive binding assay [120]

oCompetitive binding assay [121]

pCompetitive binding assay (no details given)

qCompetitive binding assay [122]

rIn-house assays

Table 2 highlights a number of important points. It demonstrates that there are very few nationally representative data; the studies are concentrated in The Gambia, Nigeria and South Africa and many studies compare control and ‘unhealthy’ groups. Table 2 also demonstrates that the 25OHD concentrations in the studies summarized should be compared and considered with caution; values were derived using 15 different methods. This leads to considerable variation in measured values for the same sample [74, 75] and the variation may even be larger when comparing older assays with modern methods. Only three papers reported the use of Quality Assurance materials and/or participation in a quality assurance scheme (e.g. the Vitamin D External Quality Assessment Scheme, www.deqas.org).

With these caveats, generally when compared to national data in the UK and US [1], Table 2 suggests that values in African countries are higher. Mean values range from 7 to 150 nmol/l and the vast majority are above 25 nmol/l. Collectively the prevalence of 25OHD <25 nmol/l averages between 5–20% and 1–5% in most age groups in the UK and US. In North African countries there is a high prevalence of low vitamin D status, 25OHD is in the rachitic range. In tropical African countries the mean 25OHD values are high, but those with rickets have lower values than their peers in the same study, even though values are generally higher than 25 nmol/l. Similarly, individuals with TB or pneumonia have lower values than control groups within the same study. Veiled women or women in purdah have a lower vitamin D status than their peers within the same country. Finally, a geographical gradient is apparent across the African continent; 25OHD is lower in North African countries and in South Africa compared with tropical African countries (i.e. higher in the latter), and within a country (different regions of South Africa).

Conclusions

Studies in Africa that have assessed 25OHD suggest a range of status from deficiency (in the rachitic range) to relatively high values. Well-established health consequences of vitamin D deficiency in Africa include rickets and osteomalacia. Emerging evidence also suggests that consequences also include increased susceptibility to infectious disease. In some African countries, risk factors for vitamin D deficiency may also include very low calcium intake, and the burden of infectious disease whereby utilisation and turnover of vitamin D is increased. Thus, a plasma 25OHD concentration considered to be sufficient in healthy people may not apply to those, for example, on very low calcium intakes, or who have tropical enteropathies, HIV or TB.

Underlying calcium nutrition must be considered when interpreting vitamin D status and considering the causes and health consequences of vitamin D deficiency in Africa. Vitamin D or calcium-deficiency rickets rather than being routinely screened for often presents as co-morbidities with infectious disease, for example pneumonia [76] and more work may give insights into the aetiology of various conditions where poor vitamin D status and/or low calcium intakes may be predisposing factors.

Studies are needed to confirm if emerging relationships between vitamin D status and health outcomes such as diabetes and cancer identified in Western countries can be replicated in African countries.

There is a need for the collection of nationally representative data of vitamin D status across Africa, with samples analysed using comparable techniques by accredited laboratories.

In summary, ‘Africa’ is not a homogenous entity with respect to geography, climate, water sources, food production and availability, or the religious and cultural practices, skin pigmentation, and burden of infectious and chronic disease of its people. All these factors are likely to affect and/or be affected by vitamin D status, and therefore how vitamin D deficiency is defined and its health consequences determined. It is therefore important to recognise that policy aspects of vitamin D cannot therefore encompass the whole continent of Africa, but need to be considered in a more local context.