Clinical Endocrinology (2016) 84, 159171

doi: 10.1111/cen.12760

REVIEW ARTICLE

Vitamin D and insulin resistance

Ian R. Wallace*,, Helen J. Wallace*,, Michelle C. McKinley, Patrick M. Bell* and Steven J. Hunter*
*Regional Centre for Endocrinology and Diabetes, Royal Victoria Hospital and Nutrition and Metabolism Group, Centre for Public
Health, Queens University Belfast, Belfast, UK

Introduction
Summary
Vitamin D is a steroid hormone, which in active form binds to
the vitamin D receptor. Expression of the vitamin D receptor in
diverse cell types (pancreatic islet cells, myocytes, hepatocytes
and adipocytes) raises the suspicion that vitamin D may be
involved in multiple cellular processes, including the response to
insulin. Insulin resistance is a characteristic feature of type 2
DM, and its attenuation may reduce the incidence of type 2 DM
and cardiovascular disease. In observational studies, low serum
25-hydroxyvitamin D (25-OHD) concentrations are associated
with an increased risk of type 2 DM. It has been suggested that
increasing serum 25-OHD concentrations may have beneficial
effects on glucose and insulin homeostasis. However, cross-sectional and interventional studies of vitamin D supplementation
provide conflicting results and demonstrate no clear beneficial
effect of vitamin D on insulin resistance. These studies are
complicated by inclusion of different patient cohorts, different
25-OHD assays and different doses and preparations of vitamin
D. Any possible association may be confounded by alterations in
PTH, 1,25-dihydroxyvitamin D or tissue vitamin D concentrations. We identified 39 studies via MEDLINE and PUBMED.
We review the evidence from 10 studies (seven observational
and three interventional) examining vitamin D and type 2 DM
incidence, and 29 studies (one prospective observational, 12
cross-sectional and 16 interventional trials) examining vitamin
D and insulin resistance. Based on this data, it is not possible to
state that vitamin D supplementation has any effect on type 2
DM incidence or on insulin resistance. Data from the multiple
ongoing randomized controlled trials of vitamin D supplementation due to report over the next few years should help to clarify
this area.
(Received 20 October 2014; returned for revision 6 November
2014; finally revised 24 February 2015; accepted 26 February 2015)

Correspondence: Dr Ian Wallace, Regional Centre for Endocrinology and

Diabetes, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BA,
UK. Tel.: 028 9063 4462; Fax: 028 9031 0111;
E-mail: iwallace01@qub.ac.uk
2015 John Wiley & Sons Ltd

Vitamin D is a steroid hormone, which in active form binds to

the vitamin D receptor. Expression of the vitamin D receptor in
diverse cell types throughout the body raises the suspicion that
vitamin D may be involved in multiple cellular processes,
including the response to insulin. Insulin resistance is defined as
a reduced biological response to insulin.1 It is strongly associated
with cardiovascular disease (CVD), is a characteristic feature of
type 2 DM,2,3 and its attenuation may reduce the incidence of
type 2 DM and CVD. In this article, we describe the physiology
of vitamin D and review evidence of an association between
serum 25-hydroxyvitamin D (25-OHD) concentration and insulin resistance.

Vitamin D physiology and mode of action

The term vitamin D refers to a group of sterols including
vitamin D2 (ergocalciferol) and vitamin D3 (cholecalciferol).
Vitamin D may be synthesized by skin cells in response to ultraviolet irradiation (UVB) or obtained from dietary sources. Plant
sources contain vitamin D2 whereas sunlight and animal sources
provide vitamin D3. UVB irradiation of the precursor molecule,
7-dehydrocholesterol, leads to formation of previtamin D. Vitamin D undergoes two hydroxylation steps, first in the liver and
subsequently in the proximal convoluted tubule of the kidney,
resulting in the active form (1,25-dihydroxyvitamin D).
The human vitamin D receptor is a 50- to 60-kDa molecule
with strong affinity for the active 1,25-dihydroxyvitamin D form.
The vitamin D receptor is present in diverse cell types. Pancreatic islet cells have both vitamin D receptors and vitamin
D-dependent calcium-binding proteins. Insulin secretion is
dependent on changes in intracellular calcium concentration,
and 1,25-dihydroxyvitamin D has been shown to regulate betacell calcium flux in vitro and in mouse models.4 Likewise, the
intracellular response to insulin can also be affected by alterations in intracellular calcium concentration.5 The vitamin D
receptor is also highly expressed in adipocytes. Activation modulates intracellular calcium concentrations and can effect changes
in lipogenesis and lipolysis.4 It is therefore plausible that vitamin
D may be involved in beta-cell secretory activity and in modulating the tissue response to insulin. It is not yet clear whether
the vitamin D receptor is expressed in myocytes and vitamin D
effects on muscle insulin sensitivity, if any, may be indirect.6,7
159

160 I. R. Wallace et al.

As vitamin D is primarily derived from UVB-induced synthesis, sunlight exposure, latitude, season and skin pigmentation are
all relevant determinants of population serum 25-OHD concentrations.8 Serum concentrations of 25-OHD decrease with age.
This may be due to decreased sun exposure; however, the skins
capacity to synthesize vitamin D is reduced in individuals aged
over 65 years to approximately 25% of that seen in individuals
aged 2030 years.9 The relationship between obesity and serum
25-OHD concentration is complex. Serum 25-OHD concentrations decrease with increasing adiposity; however, the fall in
serum 25-OHD concentrations during winter is attenuated in
obese individuals.10 In the British Isles, the major contributor to
changes in serum 25-OHD concentrations is UVB-induced synthesis, with surveys of postmenopausal British women suggesting
UVB-induced synthesis accounts for 80% of serum 25-OHD
concentrations.11
It is common to report serum 25-OHD concentrations as a
measure of vitamin D status. This is the major circulating form
of vitamin D and is widely considered to be the best indicator
of vitamin D status.12 There are a number of techniques used to
measure 25-OHD, each with strengths and weaknesses. Different
assays result in variation of up to 20% above or below those
obtained using the gold standard technique of ultra-performance liquid chromatography and tandem mass spectrometry,13
which affords measurement of both the D2 and the D3 isoforms.14

Literature search strategy

For this nonsystematic review, the databases MEDLINE and
PUBMED were searched for English-language articles through to
April 2014. Search terms included vitamin D, vitamin D2, vitamin D3, 25-OHD, 1,25-dihydroxyvitamin D, insulin resistance,
prediabetes, impaired glucose tolerance, euglycaemichyperinsulinaemic clamp, oral glucose tolerance test, OGTT, homeostasis
model assessment of insulin resistance, HOMA-IR, quantitative
insulin sensitivity check index, QUICKI, Matsuda index, hyperglycaemic clamp, glycated haemoglobin and HbA1c. Additional
publications were identified from citations of the recovered articles. We excluded studies in children and participants with type
1 DM.

Vitamin D and type 2 DM risk

Many studies have examined the association between incidence
of type 2 DM and vitamin D intake or status assessed using estimates of dietary intake and measurements of circulating serum
25-OHD concentration (Table 1a,b). Studies examining dietary
intake1517 are limited by the fact that estimates may not be
strongly correlated with serum 25-OHD concentrations, due to
differences between subjects in absorption, adiposity, sun exposure and bioavailability amongst other factors. In a systematic
review and meta-analysis including two of the three dietary estimate studies, results from the Nurses Health Study15 and the
Black Womens Health Study16 were pooled revealing an odds

ratio for type 2 DM incidence of 0
82, when comparing the

highest to lowest quartiles of combined vitamin D and calcium
intake.18
Increased consumption of low-fat dairy products and a
reduced risk of type 2 DM have been demonstrated in the Black
Womens Health Study and in women in the Japan Public
Health Centre-based Prospective Study.16,17 In the USA, all dairy
products are fortified with vitamin D leading to the suggestion
that the association between increased dairy product consumption and decreased risk of incident type 2 DM may be due, at
least partly, to vitamin D.18 However, it is possible that other
constituents of dairy products, such as magnesium, calcium,
medium-chain fatty acids, transpalmitoleic acid, casein and whey
proteins, or other not yet recognized constituents may reduce
type 2 DM risk.19
Three studies (two nested case control and one prospective
cohort)2022 have used a serum assay, and one prospective
cohort study a formula, to calculate23 serum 25-OHD concentrations. In a pooled meta-analysis of these four studies, subjects
with serum 25-OHD concentration 50 nmol/l had a 43% lower
risk of developing type 2 DM.24
Three randomized controlled trials (Table 1b) report no significant change in incidence of type 2 DM after an oral vitamin
D intervention (400800 IU daily) after 27 years followup.18,25,26 A reduced increase in fasting plasma glucose concentration and homeostasis model assessment of insulin resistance
(HOMA-IR) over 3 years is reported in subjects with prediabetes
who received vitamin D and calcium daily.18 These studies used
different dosing regimens; one had poor compliance26 and only
one specifically looked at a high-risk subgroup.18
These observational studies suggest that a lower vitamin D
status is associated with increased risk of developing type 2 DM;
however, intervention studies to date have not demonstrated a
reduction in incidence of type 2 DM in response to vitamin D
supplementation (Table 1b).

Vitamin D and insulin resistance

Insulin resistance is strongly associated with cardiovascular disease (CVD) and is a characteristic feature of type 2 DM.2,3 If
serum 25-OHD concentration is related to type 2 DM risk, this
may be mediated via an effect on insulin resistance. It is attractive in research studies to use insulin resistance, which may
change rapidly in response to an intervention, as a surrogate
end-point for clinical outcomes which may not develop for
many years. It is not yet proved that attenuation of insulin resistance leads to a reduction in incidence of type 2 DM and CVD,
raising the possibility that studies conducted to investigate the
effects on insulin resistance are of limited clinical utility.
A variety of methods are used to assess insulin resistance.27 A
number of cross-sectional and interventional studies have examined the relationship between vitamin D and insulin resistance
and are summarized in Tables 2 and 3, respectively.
Only one study (the Medical Research Council Ely Prospective
Study) has examined the prospective association between serum

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

Prospective
cohort

Nested case
control

Prospective
cohort

Prospective
cohort

Nested case
control

Knekt, Finland20

Liu, USA23

Pittas (2006),
USA15

Pittas (2010),
USA21

Prospective
cohort

Study design

Kirii, Japan17

(a)
Anderson, USA22

First author,
country
(reference)

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171
Women

Women

Men (46%)
and Women

Men (48%)
and Women

Men (43%)
and Women

Men (25%)
and Women

Subject sex

56
4

46

54
2

4074

4069

55

Age
(mean or
range)
(years)

Nondiabetic
nurses

Nondiabetic
nurses (98%
Caucasian)

Nondiabetic
(mostly
Caucasian)

Nondiabetic

Non-diabetic
(mostly
Caucasian)
Nondiabetic

Baseline
characteristics

1 157 (608)

83 779 (4 843)

3 066 (133)

7 503 (412)

59 796 (1 114)

41 504 (9 389)

Number of
subjects (number
of cases)

14 years

20 years

7 years
(mean)

1722 years

5 years

1
3 years
(mean)

Period of
follow-up

Type 2 DM
(validated selfreport)

Fasting plasma
glucose
7
0 mmol/l or
use of
hypoglycaemic
agents
Type 2 DM
(validated selfreport)

Type 2 DM (from
National health
register
database)

Type 2 DM
(Health register
database)
DM (Type not
specified)
(validated selfreport)

Outcome measure

Table 1. (a) Vitamin D and type 2 DM risk (observational studies). (b) Vitamin D and type 2 DM risk (randomized controlled trials)

Serum

Dietary
intake

Calculation

Serum

Dietary
intake

Serum

Exposure:
Vitamin D
intake or
status

Vit D 800 IU/day

and Calcium
1200 mg/day
compared to Vit D
400 IU/day and
Calcium 600 mg/
day
Highest compared to
lowest 25-OHD
quartile

412 cases compared

to 986 matched
controls. Highest
compared to lowest
vitamin D quartile
Highest compared to
lowest third

Highest quartile of
calcium intake
compared to lowest
quartile of calcium
intake in those with
vitamin D intake
greater than the
median intake

Lowest compared to
highest third

Predictor (Groups
compared)

OR: 0
52 95% CI

(0
330
83)

RR: 0
67 95% CI

(0
490
90)

HR: 0
60 95% CI

(0
37 0
97)

Men: RR: 0
62 95%

CI (0
410
94)
Women: RR: 0
59
95% CI (0
380
91)
(calcium intake and
vitamin D intake
alone were not
associated with DM
incidence in men or
women)
OR: 0
60 95% CI
(0
37 0
96)

HR: 1
89 95% CI

(1
54 2
33)

Main study results

Vitamin D and insulin resistance 161

71
2

62

77

Age
(mean
or range)
(years)

Women

Subject sex

Nondiabetic
(29%
prediabetes)

Nondiabetic
(84%
Caucasian)

Nondiabetic

Baseline
characteristics

39

Age
(mean or
range)
(years)

314 (case
numbers not
reported)

33 951 (2
291)

5 292 (104)

Number of
subjects
(number of
cases)

Nondiabetic
(African
American)

Baseline
characteristics

3 years

7 years
(median)

2462
months

Period of
follow-up

41 186 (1 964)

8 years
(maximum)

Period of
follow-up

Type 2 DM or
prediabetes
(fasting plasma
glucose
measurement)

Type 2 DM
(treatment
with
hypoglycaemic
agents)

Type 2 DM
(self-report)

Outcome
measure

Number of
subjects (number
of cases)

76 (mean)

38 (mean)
25-OHD only
reported for
subsample of 60
participants in
intervention
group
43
7 (median)

Baseline 25-OHD
concentration
(nmol/l)

Type 2 DM
(validated selfreport)

Outcome measure

Placebo group
73 (mean)
Intervention
group 107
(mean)

62 (mean)
25-OHD only
reported for
subsample of 60
participants in
intervention
group
NR

Compared
intervention
group (1000 mg
calcium and
800 IU vitamin
D daily) to
placebo group
Compared
intervention
group (1000 mg
calcium and
400 IU vitamin
D daily) to
placebo group
Compared
intervention
group (500 mg
calcium and 700
IU vitamin D
daily) to placebo
group

Progression to
type 2 DM or
prediabetes 19%
in placebo group
and 20% in
intervention
group
(P = 0
84)

HR: 1
01 95% CI

(0
941
10)

OR: 1
11 95% CI

(0
771
62)

Main study
results

RR: 0
86 95% CI

(0
74 1
00) (This
was no longer
significant after
correction for
magnesium intake)

Main study results

Predictor (Groups
compared)

Highest quintile
(Calcium 661 mg/
day) compared to
lowest quintile
(Calcium 219 mg/
day)

Predictor (Groups
compared)

Postintervention
25-OHD
concentration
(nmol/l)

Dietary
intake

Exposure:
Vitamin D
intake or
status

DM, diabetes mellitus; RR, relative risk; CI, confidence interval; OR, odds ratio; HR, hazard ratio; 25-OHD, 25-hydroxyvitamin D; NR, not reported.

Men (42%)
and Women

Women

de Boer, USA25

Pittas (2007),
USA18

Men (15%)
and Women

Subject sex

Prospective
cohort

Study design

(b)
Avenell, UK
(England and
Scotland) 26

First author,
country
(reference)

van Dam, USA16

First author,
country
(reference)

Table 1. (continued)

162 I. R. Wallace et al.

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

 2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

Men (54%) and

Women

Men (42%) and

Women

Men (42%) and

Women
Men (53%) and
Women

Men (50%) and

Women

Men (30%) and

Women
Women

Women

Men (50%) and

Women

Men (36%) and

Women
Men (49%) and
Women

Rajakumar, USA30

de las Heras, USA31

Chiu, USA32

Liu, USA34

Kayaniyil, Canada35

Maghbooli, Iran37

Delvin, Canada38

Jorde, Norway39
Not
reported

2170

916

Not
reported

14

49
6

59
6

62

26

14
3

12
6

41
4

Children and
adolescents (FrenchCanadian)
Obese and
overweight adults
US adults without
physician diagnosed
DM

Obese adolescents
(African American)
Pregnant women

Nondiabetic

Normal glucose
tolerance
15 Secondary
hyperparathyroidism
and 15 normal
controls
Nondiabetic

US youths (Mixed
glucose tolerance)

Mixed glucose
tolerance (7 IGT
and 1 type 2 DM)
US youths
(nondiabetic)

Baseline
characteristics

NR

58 (mean)

46 (mean)

54 (mean)

37 (mean)

56 (mean)

3206 (118 with

undiagnosed
DM)

438

1745

741 (52 with

gestational DM)

51

712

808

30

58 in 50%

47 (mean)

126

175

57 (mean)

44 (mean)

183

39

<50 in 54%

51 (mean)

Number of
subjects

25-OHD
concentration
(nmol/l)

HOMA-IR

HOMA-IR and QUICKI

HOMA-IR

HOMA-IR and Matsuda

Index
HOMA-IR and Matsuda
Index
HOMA-IR,- QUICKI- and
OGTT-derived ISI

HOMA-IR- and OGTTderived ISI

EHC-, OGTT- and

hyperglycaemic clampderived indices
Hyperglycaemic clampderived indices
OGTT- and hyperglycaemic
clamp-derived indices

EHC- and hyperglycaemic

clamp-derived indices

EHC-derived indices

Technique of assessment of
insulin resistance

HOMA-IR 12
7% lower in highest

compared to lowest vitamin D tertile.
No association with ISI after
correction for BMI and waist
circumference
HOMA-IR r = 0
29 (P < 0
01)
Matsuda Index r = 0
30 (P < 0
01)
No significant correlation coefficients in
total group
HOMA-IR r = 0
2 (P < 0
01)
QUICKI not significant
ISI r = 0
46 (P = 0
01)
Reduction of 2
8% (males) and 2
3%
(females) for each 10 nmol/l increase
in vitamin D concentration
HOMA-IR r = 0
12 (P < 0
05)
QUICKI r = 0
12 (P < 0
05)
Linear decrease in HOMA-IR with
increase in vitamin D concentration

No association

Clamp-derived insulin sensitivity

r = 0
173 (P = 0
019). This was no
longer significant after adjustment for
adiposity
No significant association in normal
glucose tolerance, impaired glucose
tolerance or type 2 DM subgroups
ISI r = 0
46 (P < 0
01)

GIR r = 0
43 (P < 0
01)

Association between vitamin D

concentration and outcome measure of
insulin sensitivity

DM, diabetes mellitus; EHC, euglycaemichyperinsulinaemic clamp; GIR, glucose infusion rate; IGT, impaired glucose tolerance; ISI, insulin sensitivity index; HOMA-IR, homeostasis model assessment
of insulin resistance; QUICKI, quantitative insulin sensitivity check index; OGTT, oral glucose tolerance test; HbA1c, glycated haemoglobin; 25-OHD, 25-hydroxyvitamin D; NR, not reported.

Zhao, USA40

Ashraf, USA36

Kamycheva,
Norway33

Men (46%) and

Women

Subject Sex

Muscogiuri, Italy29

First author, country

(reference)

Age
(mean or
range)
(years)

Table 2. Cross-sectional studies of vitamin D and insulin resistance

Vitamin D and insulin resistance 163

Vitamin D and insulin resistance 169

incidence or on insulin resistance, a surrogate marker whose
attenuation may not be clinically relevant. The multiple ongoing
randomized controlled trials of vitamin D supplementation due
to report over the next few years may confirm this.

Conflict of interest
Nothing to declare.

Financial disclosure
Nothing to declare.

References
1 Kahn, C.R. (1978) Insulin resistance, insulin insensitivity, and
insulin unresponsiveness: a necessary distinction. Metabolism:
Clinical & Experimental, 27, 18931902.
2 Wilson, P.W., DAgostino, R.B., Parise, H. et al. (2005) Metabolic syndrome as a precursor of cardiovascular disease and type
2 diabetes mellitus. Circulation, 112, 30663072.
3 Stumvoll, M., Goldstein, B.J. & van Haeften, T.W. (2005) Type 2
diabetes: principles of pathogenesis and therapy. Lancet, 365,
13331346.
4 Shi, H., Norman, A.W., Okamura, W.H. et al. (2002)
1alpha,25-dihydroxyvitamin D3 inhibits uncoupling protein 2
expression in human adipocytes. FASEB Journal, 16, 18081810.
5 Draznin, B., Lewis, D., Houlder, N. et al. (1989) Mechanism of
insulin resistance induced by sustained levels of cytosolic free
calcium in rat adipocytes. Endocrinology, 125, 23412349.
6 Wang, Y. & DeLuca, H.F. (2011) Is the vitamin d receptor found
in muscle? Endocrinology, 152, 354363.
7 Ceglia, L. & Harris, S.S. (2013) Vitamin D and its role in skeletal
muscle. Calcified Tissue International, 92, 151162.
8 McKenna, M.J. (1992) Differences in vitamin D status between
countries in young adults and the elderly. American Journal of
Medicine, 93, 6977.
9 Need, A.G., Morris, H.A., Horowitz, M. et al. (1993) Effects
of skin thickness, age, body fat, and sunlight on serum
25-hydroxyvitamin D. American Journal of Clinical Nutrition, 58,
882885.
10 Forsythe, L.K., Livingstone, M.B., Barnes, M.S. et al. (2012)
Effect of adiposity on vitamin D status and the 25-hydroxycholecalciferol response to supplementation in healthy young and
older Irish adults. British Journal of Nutrition, 107, 126134.
11 Macdonald, H.M., Mavroeidi, A., Fraser, W.D. et al. (2011) Sunlight and dietary contributions to the seasonal vitamin D status
of cohorts of healthy postmenopausal women living at northerly
latitudes: a major cause for concern? Osteoporosis International,
22, 24612472.
12 Seamans, K.M. & Cashman, K.D. (2009) Existing and potentially
novel functional markers of vitamin D status: a systematic
review. American Journal of Clinical Nutrition, 89, 1997S2008S.
13 Bouillon, R. (2012) Report on recent vitamin D research: ECTS
2012 and the 15th vitamin D workshop 2012. International Bone
and Mineral Society Bone Key Reports, 9, 203.
14 Tahsin-Swafiri, S., Bianco-Navarro, I., Perez-Sacristan, B. et al.
(2012) The prevalence of vitamin deficiency in clinical practice is
assay-dependent. Clinical Nutrition, 31, 10111014.

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

15 Pittas, A.G., Dawson-Hughes, B., Li, T. et al. (2006) Vitamin D

and calcium intake in relation to type 2 diabetes in women. Diabetes Care, 29, 650656.
16 van Dam, R.M., Hu, F.B., Rosenberg, L. et al. (2006) Dietary calcium and magnesium, major food sources, and risk of type 2
diabetes in U.S. black women. Diabetes Care, 29, 22382243.
17 Kirii, K., Mizoue, T., Iso, H. et al. (2009) Calcium, vitamin D
and dairy intake in relation to type 2 diabetes risk in a Japanese
cohort. Diabetologia, 52, 25422550.
18 Pittas, A.G., Harris, S.S., Stark, P.C. et al. (2007) The effects of
calcium and vitamin D supplementation on blood glucose and
markers of inflammation in nondiabetic adults. Diabetes Care,
30, 980986.
19 Lacroix, I.M.E. & Li-Chan, E.C.Y. (2014) Investigation of the
putative associations between dairy consumption and incidence
of type 1 and type 2 diabetes. Critical Reviews in Food Science
and Nutrition, 54, 411432.
20 Knekt, P., Laaksonen, M., Mattila, C. et al. (2008) Serum vitamin D and subsequent occurrence of type 2 diabetes. Epidemiology, 19, 666671.
21 Pittas, A.G., Sun, Q., Manson, J.E. et al. (2010) Plasma 25hydroxyvitamin D concentration and risk of incident type 2
diabetes in women. Diabetes Care, 33, 20212023.
22 Anderson, J.L., May, H.T., Horne, B.D. et al. (2010) Relation of
vitamin D deficiency to cardiovascular risk factors, disease status,
and incident events in a general healthcare population. The
American Journal of Cardiology, 106, 963968.
23 Liu, E., Meigs, J.B., Pittas, A.G. et al. (2010) Predicted 25hydroxyvitamin D score and incident type 2 diabetes in the
Framingham Offspring Study. American Journal of Clinical Nutrition, 91, 16271633.
24 Mitri, J., Muraru, M.D. & Pittas, A.G. (2011) Vitamin D and
type 2 diabetes: a systematic review. European Journal of Clinical
Nutrition, 65, 10051015.
25 de Boer, I.H., Tinker, L.F., Connelly, S. et al. (2008) Calcium
plus vitamin D supplementation and the risk of incident diabetes
in the Womens Health Initiative. Diabetes Care, 31, 701707.
26 Avenell, A., Cook, J.A., MacLennan, G.S. et al. (2009) Vitamin D
supplementation and type 2 diabetes: a substudy of a randomised placebo-controlled trial in older people (RECORD trial,
ISRCTN 51647438). Age and Ageing, 38, 606609.
27 Muniyappa, R., Lee, S., Chen, H. et al. (2008) Current
approaches for assessing insulin sensitivity and resistance in vivo:
advantages, limitations, and appropriate usage. American Journal
of Physiology Endocrinology & Metabolism, 294, E15E26.
28 Forouhi, N.G., Luan, J., Cooper, A. et al. (2008) Baseline serum
25-hydroxy vitamin d is predictive of future glycemic status and
insulin resistance: the Medical Research Council Ely Prospective
Study 1990-2000. Diabetes, 57, 26192625.
29 Muscogiuri, G., Sorice, G.P., Prioletta, A. et al. (2010) 25Hydroxyvitamin D concentration correlates with insulinsensitivity and BMI in obesity. Obesity, 18, 19061910.
30 Rajakumar, K., de las Heras, J., Lee, S. et al. (2012) 25-hydroxyvitamin D concentrations and in vivo insulin sensitivity and
beta-cell function relative to insulin sensitivity in black and
white youth. Diabetes Care, 35, 627633.
31 de las Heras, J., Rajakumar, K., Lee, S. et al. (2013) 25-Hydroxyvitamin D in obese youth across the spectrum of glucose tolerance from normal to prediabetes to type 2 diabetes. Diabetes
Care, 36, 20482053.

Randomize
controlled trial
(double blind)

Randomized
controlled trial
(double blind)

Randomized
controlled
trial, crossover
(double blind)

Randomized
controlled
trial, parallel
(double blind)

Controlled trial

Randomized
controlled
trial, parallel
(double blind)

Jorde (2009),
Norway49

Orwoll, USA41

Sugden, UK
(Scotland)51

Borissova,
Bulgaria50

Witham, UK
(Scotland)52

Study design

Davidson, USA48

First author
(reference)

Table 3. (continued)

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

Oral vitamin D2
100 000 IU or
200 000 IU

Oral vitamin D3
1332 IU daily

Oral vitamin D2
100 000 IU

Oral 1,25dihydroxyvitamin D
1 lg daily

Intramuscular D3
once weekly (dose
based on subject
weight and baseline
vitamin D
concentration)
(mean weekly dose
88 865  16 154
IU)
Oral vitamin D3
40 000 IU weekly

Intervention

Women
(postmenopausal)

Men (69%) and

Women

Single dose

Men (53%) and

Women

Men (% not
reported) and
Women

Men (56%) and

Women

Men (32%) and

women

Subject sex

1 month

Single dose

4 days

6 months

12 months

Duration of
intervention

65
1

53
8

64
2

4070

2175

52
4

Age
(mean or
range)
(years)

Type 2 DM and
vitamin D
<100 nmol/l

Type 2 DM

Type 2 DM and
vitamin D
<50 nmol/l

Type 2 DM

Type 2 DM (on
metformin and one
long-acting insulin
injection per day)

Prediabetes and
serum vitamin D
concentration
75 nmol/l

Subject
characteristics

28 (11 in
intervention
group and 17
controls)
61 (19 in 100 000
IU group, 20 in
200 000 IU
group and 22 in
placebo group)

34 (17 in
intervention
group and 17 in
placebo group)

36 (32 completed
study, 16 in
intervention
group and 16 in
placebo group)
20

109 (56 in
intervention
group and 53 in
placebo group)

Number of
subjects

63 (mean)

79 (mean) in
20 000 IU group,
63 (mean) in
10 000 IU group,
54 (mean) in
placebo group

NR (>50 in 70%)

NR
Report 1,25dihyroxyvitamin
D concentration
rise from 82
103 pmol/l in
intervention
group and 82
91 pmol/l in
placebo group
63 (mean) in
intervention
group, 43 (mean)
in placebo group

118 (mean) in
intervention
group, 57 (mean)
in placebo group

Reported as nearly
175 (mean) in
intervention
group and no
change in
placebo group

Postintervention
25-OHD
concentration
(nmol/l)

35 (mean)

38 (mean)

35 (mean)

59 (mean)

55 (mean)

Baseline 25-OHD
concentration
(nmol/l)

Serum

Serum

HOMA-IR

Serum

Serum

Serum

Serum

Method of
assessment
of vitamin D

HOMA-IR

HOMA-IR

SUSTACAL
stimulation test

HOMA-IR, Fasting
glucose, Fasting
insulin, HbA1c

HOMA-IR,
Matsuda index

Technique of
assessment of
insulin resistance

No significant
change

No significant
change in whole
group. In
subjects who had
an increase in
vitamin D
>11 nmol/l,
HOMA-IR was
significantly
reduced
No significant
change in
HOMA-IR

No significant
change

No significant
change

No significant
change in
HOMA-IR or
Matsuda index

Results

Vitamin D and insulin resistance 165

Controlled trial

Controlled trial

Mak (1998),
USA54

Kautzky-Willer,
Austria55

IV 1,25dihydroxyvitamin D
(mean dose
0
96  0
08 lg) 3
times per week

IV 1,25dihydroxyvitamin
D3 daily

IV 1,25dihydroxyvitamin
D3 (2 lg/m2)

Intervention

12 weeks

4 weeks

Single dose

Duration of
intervention

Men (% not
reported) and
Women

Men (% not
reported) and
Women

Men (% not
reported) and
Women

Subject sex

39

19

1623

Age
(mean or
range)
(years)

Chronic renal failure,

on haemodialysis

Chronic renal failure,

on haemodialysis

Chronic renal failure,

on haemodialysis

Subject
characteristics

20 (10 in
intervention
group and 10
healthy controls)

23 (8 in
intervention
group, 8 in
control group on
haemodialysis
and 7 healthy
controls not on
haemodialysis)

22 (11 in
intervention
group and 11 in
placebo group)

Number of
subjects

NR
Report 1,25dihyroxyvitamin
D of 44 pmol/l
(mean) in
intervention
group, 36 pmol/l
(mean) in
haemodialysis
controls and not
reported in
healthy control
group
NR
Report 1,25dihyroxyvitamin
D of 63 pmol/l
(mean) in
intervention
group, not
reported in
healthy control
group

NR
Report 1,25dihyroxyvitamin
D of 20 pmol/l
(mean) in
haemodialysis
group and
39 pmol/l
(mean) in
control group

Baseline 25-OHD
concentration
(nmol/l)
NR
Report 1,25dihyroxyvitamin
D concentration
of 191 pmol/l
(mean) in
haemodialysis
group and
170 pmol/l
(mean) in
control group
NR
Report 1,25dihyroxyvitamin
D of 132 pmol/l
(mean) in
intervention
group, 40 pmol/l
(mean) in
haemodialysis
controls and not
reported in
healthy control
group
NR
Report 1,25dihyroxyvitamin
D of 145 pmol/l
(mean) in
intervention
group, not
reported in
healthy control
group

Postintervention
25-OHD
concentration
(nmol/l)

Serum

EHC-derived
indices

Serum

Serum

FSIVGGT

FSIVGGT

Method of
assessment
of vitamin D

Technique of
assessment of
insulin resistance

Increased ISI to
levels seen in the
nondialysis
control group

Significant
improvement in
glucose disposal
rate

Improved glucose
uptake to levels
seen in controls

Results

DM, diabetes mellitus; EHC, euglycaemichyperinsulinaemic clamp; FSIVGTT, frequently sampled intravenous glucose tolerance test; ISI, insulin sensitivity index; HOMA-IR, homeostasis model assessment of insulin resistance; QUICKI, quantitative insulin sensitivity check index; OGTT, oral glucose tolerance test; HbA1c, glycated haemoglobin; OGIS, oral glucose insulin sensitivity index; HOMA2IR, revised homeostasis model assessment of insulin resistance; 25-OHD, 25-hydroxyvitamin D; NR, not reported.

Controlled trial

Study design

Mak (1992),
USA53

First author
(reference)

Table 3. (continued)

166 I. R. Wallace et al.

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

Vitamin D and insulin resistance 167

25-OHD concentration and insulin resistance. After 10 years,
baseline serum 25-OHD concentration was inversely associated
with fasting plasma glucose concentration, 2-h plasma glucose
concentration on oral glucose tolerance test (OGTT), fasting
plasma insulin concentration and HOMA-IR.28

Vitamin D and insulin resistance: cross-sectional

studies
Three cross-sectional studies have used the euglycaemichyperinsulinaemic clamp technique2931; two, the hyperglycaemic
clamp32,33; five, OGTT-derived indices31,3336; and seven studies
report HOMA-IR or quantitative insulin sensitivity check index
(QUICKI)3440 to assess insulin resistance (Table 2).
Studies using the gold standard euglycaemichyperinsulinaemic clamp technique report no association after controlling for
BMI and no difference in serum 25-OHD concentration or insulin sensitivity amongst subgroups with mixed glucose tolerance
(normal glucose tolerance, impaired glucose tolerance and type
2 DM).2931
Cross-sectional studies provide conflicting results with two
studies demonstrating no association between serum 25-OHD
concentration and indices of insulin resistance,31,33 and the
remainder reporting varying degrees of association.29,30,32,3440
After correction for adiposity, associations were no longer significant in three studies.29,30,34 Studies included subjects with a spectrum of glucose tolerance,29,31 gestational diabetes,37 type 2 DM41
and secondary hyperparathyroidism.33 It is difficult to draw any
meaningful conclusions from this mixture of cross-sectional studies, and further investigation of defined cohorts is required.

Vitamin D and insulin resistance: intervention

studies
Sixteen intervention trials examining the effects of vitamin D on
insulin resistance are summarized in Table 3.
In nondiabetic cohorts, an improvement in insulin resistance
(assessed by HOMA2-IR) has been demonstrated in one randomized controlled double-blind trial of women of South Asian
ethnicity following vitamin D supplementation for 6 months.42
In contrast, other intervention studies in nondiabetic cohorts
have demonstrated no effect of vitamin D supplementation on
insulin resistance.39,4346 These studies included hyperinsulinaemiceuglycaemic clamp-derived indices of insulin sensitivity,43
HOMA-IR and QUICKI,39 HOMA-IR,44 fasting plasma glucose
concentration and HbA1c45 and fasting plasma glucose concentration46 as outcome measures.
Only two studies have examined the effect of a vitamin D
intervention on insulin resistance in patients with prediabetes or
impaired glucose tolerance, with both demonstrating no benefit.47,48 A small study of 14 middle-aged Swedish men with
impaired glucose tolerance is limited by size and a suboptimal
protocol.47 In a randomized controlled double-blind trial, no
change in insulin resistance (assessed by HOMA-IR and Matsuda index) and progression to type 2 DM between groups was
evident.48
2015 John Wiley & Sons Ltd
Clinical Endocrinology (2016), 84, 159171

In patients with type 2 DM, five studies have demonstrated

no effect of vitamin D administration on insulin resistance.41,49
52
Sugden et al. report two studies with contrasting results on
the primary outcome measure of endothelial function. In the
first, subjects with type 2 DM and a serum 25-OHD concentration 50 nmol/l were examined, demonstrating no difference in
HbA1c or HOMA-IR between the groups. However, on subgroup analysis, a statistically significant reduction in HOMA-IR
was demonstrated in the subjects who had an increase in serum
25-OHD concentration >11 nmol/l. The numbers achieving this
increase in serum 25-OHD concentration were not reported,
must have been small, and HOMA-IR was a secondary outcome
measure. Endothelial function (the primary outcome measure)
did improve, with an increase in flow-mediated dilatation of the
brachial artery.51 The same group has also reported the effects of
differing doses of vitamin D on endothelial function in a study
including subjects with type 2 DM and a serum 25-OHD concentration <100 nmol/l, demonstrating no change in HbA1c,
fasting plasma glucose concentration or HOMA-IR, and in contrast to their previous study, no significant effect on endothelial
function was demonstrated.52 Participants in the previous study
had a lower baseline serum 25-OHD concentration, suggesting a
possible threshold for effects of vitamin D on endothelial function.51
Three studies in subjects with chronic renal failure undergoing
haemodialysis have demonstrated a reduction in insulin resistance in response to vitamin D.5355 These studies are all small,
of short duration, and results may not be applicable to the nondialysis population.
Intervention studies have utilized a variety of techniques to
assess insulin resistance. Only three have used the euglycaemic
hyperinsulinaemic clamp technique43,47,54 each in different subject groups and one with a suboptimal study design.47 Vitamin
D has been administered orally or intravenously, in a variety of
doses, for variable duration and in different formulations. Only
one study in a nondiabetic cohort has reported an improvement
in insulin resistance.42 The South Asian (mostly Indian) women
in this study were insulin resistant (HOMA-IR > 1
93) and had
a rise in serum 25-OHD concentration to a median of 80 nmol/
l.42 In Indian men with central adiposity, no beneficial effect on
HOMA-IR was demonstrated44; however, the intervention in the
women,42 who were now resident in an area of lower sun exposure, lasted for 6 months compared to only 6 weeks in the men,
suggesting that duration of treatment may be important. In
many other intervention studies, baseline and postintervention
characteristics are less clearly reported.
In subjects with type 2 DM, no effect on HOMA-IR5052 or
on meal testing indices has been demonstrated.41 In subjects
with chronic kidney disease and serum 25-OHD concentrations
3043 nmol/l, intravenous vitamin D improves insulin secretion
and insulin sensitivity, without change in plasma PTH concentration.5355 Only one study has examined a cohort with
impaired glucose tolerance, demonstrating no effect of alphacalcidol over 18 months on insulin sensitivity.47 No effect of a
1-year intervention has been demonstrated in individuals with
prediabetes.48

168 I. R. Wallace et al.

In summary, intervention studies, to date, provide largely negative results. Only one study in a nondiabetic, nonchronic kidney disease cohort has reported an improvement in insulin
resistance. It is not possible from this data to suggest that vitamin D and insulin resistance are causally linked.

Ongoing studies
There are presently over 150 ongoing trials of vitamin D supplementation involving type 2 DM, cardiovascular disease and insulin resistance outcomes, registered with the World Health
Organisation International Clinical Trials Registry. We draw
attention to only a small proportion of these ongoing trials, with
three randomized controlled trials examining the effect of vitamin D supplementation on incidence of type 2 DM in cohorts
at increased risk of type 2 DM with follow-up over a number of
years.5658 We also draw attention to three randomized controlled trials examining the effects of vitamin D supplementation
on insulin resistance in overweight and obese individuals and
individuals with prediabetes, as we believe these are welldesigned trials in high-risk cohorts.5961 All are due to report in
20152017. It is hoped that these will assist in definitively
addressing the potential role of vitamin D in reducing insulin
resistance and type 2 DM.

Issues to consider in design of future research

Optimal serum 25-OHD concentrations have not been defined,
and it is possible that these concentrations differ for various
outcomes. A u-shaped relationship, with increased risk for lowand high-serum 25-OHD concentrations has been reported for
cancer, cardiovascular events and mortality. It is possible that
studies have not achieved adequate serum 25-OHD concentrations to influence definitive outcomes.62 Furthermore, different
preparations of vitamin D may have different effects, with a
recent meta-analysis reporting reduced mortality in patients
given vitamin D3, but not vitamin D2.63
Polymorphisms of the vitamin D receptor gene are associated
with a decreased rise in serum 25-OHD concentration in
response to vitamin D supplementation, allowing identification
of low responders.64 Metabolomic phenotyping may identify
subjects who could benefit from vitamin D supplementation.
Such a phenotype was identified in a randomized double-blind
controlled study of 4 weeks of vitamin D supplementation, demonstrating reduced fasting plasma insulin concentration and
reduced HOMA-IR.65
Inflammation has been proposed as a potential link between
vitamin D deficiency and increased risk of chronic nonskeletal
diseases, with the suggestion that inflammation causes the vitamin D deficiency. The authors of a recent systematic review suggest that vitamin D deficiency is a surrogate marker of systemic
inflammation rather than causing it.66
Circulating vitamin D (vitamin D, 25, OHD and 1,25-dihydroxyvitamin D) is principally bound to vitamin D-binding
protein, and indeed, vitamin D-binding protein concentration

was correlated with insulin resistance (as assessed by HOMA-IR

and QUICKI).67 In a study of postmenarcheal female adolescents, it was observed that as total serum 25-OHD concentration
falls, so too does vitamin D-binding protein concentration,
maintaining similar free (or unbound) 25-OHD concentrations.
This may be an adaptive response or it is possible that vitamin
D-binding protein itself may have a direct association with insulin resistance. The authors suggest that free serum 25-OHD
concentration may be more relevant to clinical outcomes than
total serum 25-OHD concentration and that a low vitamin
D-binding protein concentration may be the first stage of
development of vitamin D deficiency.67

Potential confounders of an association between

vitamin D and insulin resistance
Serum 25-OHD concentrations are inversely correlated with
plasma PTH concentrations,68 which regulate the activity of
renal 1-a-hydroxylase. Increases in serum 25-OHD concentration lead to a reduction in plasma PTH concentration, however,
even with reduced renal 1-a-hydroxylase activity, there may be
an increased serum concentration of the active 1,25-dihydroxyvitamin D form such that the relationship between PTH, vitamin
D, 25-OHD and active 1,25-dihydroxyvitamin D is complex.68
The association between plasma PTH concentration and insulin
resistance has been assessed in subjects with chronic kidney disease, who have secondary hyperparathyroidism and vitamin D
deficiency. Three studies demonstrate an improvement in insulin
secretion and insulin sensitivity in response to intravenous vitamin D.5355 One study demonstrated this improvement following a 4-week intervention with no change in plasma PTH
concentration, suggesting the improvement is independent of
PTH.54 It is not clear whether this is due to a pharmacological
effect or to correcting vitamin D deficiency or if this finding is
generalizable to nonchronic kidney disease subjects.
Serum 25-OHD concentrations are reported as a measure of
vitamin D status,12 but may not represent concentrations of the
active form (1,25-dihydroxyvitamin D). It is possible that there
is an association between 1,25-dihydroxyvitamin D and insulin
resistance or between tissue concentrations of vitamin D and
insulin resistance that are not reflected by serum 25-OHD concentrations.

Conclusions
Overall, whilst observational evidence suggests that vitamin D
status may be associated with T2DM, the relationship with insulin resistance is less clear. Intervention studies, to date, are generally negative. This finding is replicated in other reviews.66,69
The enthusiasm for vitamin D as a panacea for multiple health
outcomes is waning, with a recent trial sequential meta-analysis
reporting no effect of vitamin D supplementation on skeletal,
vascular or cancer outcomes, and suggesting that further studies
are unlikely to change these conclusions.70 It is possible that
vitamin D supplementation has no effect on type 2 DM

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

Vitamin D and insulin resistance 169

incidence or on insulin resistance, a surrogate marker whose
attenuation may not be clinically relevant. The multiple ongoing
randomized controlled trials of vitamin D supplementation due
to report over the next few years may confirm this.

Conflict of interest
Nothing to declare.

Financial disclosure
Nothing to declare.

References
1 Kahn, C.R. (1978) Insulin resistance, insulin insensitivity, and
insulin unresponsiveness: a necessary distinction. Metabolism:
Clinical & Experimental, 27, 18931902.
2 Wilson, P.W., DAgostino, R.B., Parise, H. et al. (2005) Metabolic syndrome as a precursor of cardiovascular disease and type
2 diabetes mellitus. Circulation, 112, 30663072.
3 Stumvoll, M., Goldstein, B.J. & van Haeften, T.W. (2005) Type 2
diabetes: principles of pathogenesis and therapy. Lancet, 365,
13331346.
4 Shi, H., Norman, A.W., Okamura, W.H. et al. (2002)
1alpha,25-dihydroxyvitamin D3 inhibits uncoupling protein 2
expression in human adipocytes. FASEB Journal, 16, 18081810.
5 Draznin, B., Lewis, D., Houlder, N. et al. (1989) Mechanism of
insulin resistance induced by sustained levels of cytosolic free
calcium in rat adipocytes. Endocrinology, 125, 23412349.
6 Wang, Y. & DeLuca, H.F. (2011) Is the vitamin d receptor found
in muscle? Endocrinology, 152, 354363.
7 Ceglia, L. & Harris, S.S. (2013) Vitamin D and its role in skeletal
muscle. Calcified Tissue International, 92, 151162.
8 McKenna, M.J. (1992) Differences in vitamin D status between
countries in young adults and the elderly. American Journal of
Medicine, 93, 6977.
9 Need, A.G., Morris, H.A., Horowitz, M. et al. (1993) Effects
of skin thickness, age, body fat, and sunlight on serum
25-hydroxyvitamin D. American Journal of Clinical Nutrition, 58,
882885.
10 Forsythe, L.K., Livingstone, M.B., Barnes, M.S. et al. (2012)
Effect of adiposity on vitamin D status and the 25-hydroxycholecalciferol response to supplementation in healthy young and
older Irish adults. British Journal of Nutrition, 107, 126134.
11 Macdonald, H.M., Mavroeidi, A., Fraser, W.D. et al. (2011) Sunlight and dietary contributions to the seasonal vitamin D status
of cohorts of healthy postmenopausal women living at northerly
latitudes: a major cause for concern? Osteoporosis International,
22, 24612472.
12 Seamans, K.M. & Cashman, K.D. (2009) Existing and potentially
novel functional markers of vitamin D status: a systematic
review. American Journal of Clinical Nutrition, 89, 1997S2008S.
13 Bouillon, R. (2012) Report on recent vitamin D research: ECTS
2012 and the 15th vitamin D workshop 2012. International Bone
and Mineral Society Bone Key Reports, 9, 203.
14 Tahsin-Swafiri, S., Bianco-Navarro, I., Perez-Sacristan, B. et al.
(2012) The prevalence of vitamin deficiency in clinical practice is
assay-dependent. Clinical Nutrition, 31, 10111014.

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

15 Pittas, A.G., Dawson-Hughes, B., Li, T. et al. (2006) Vitamin D

and calcium intake in relation to type 2 diabetes in women. Diabetes Care, 29, 650656.
16 van Dam, R.M., Hu, F.B., Rosenberg, L. et al. (2006) Dietary calcium and magnesium, major food sources, and risk of type 2
diabetes in U.S. black women. Diabetes Care, 29, 22382243.
17 Kirii, K., Mizoue, T., Iso, H. et al. (2009) Calcium, vitamin D
and dairy intake in relation to type 2 diabetes risk in a Japanese
cohort. Diabetologia, 52, 25422550.
18 Pittas, A.G., Harris, S.S., Stark, P.C. et al. (2007) The effects of
calcium and vitamin D supplementation on blood glucose and
markers of inflammation in nondiabetic adults. Diabetes Care,
30, 980986.
19 Lacroix, I.M.E. & Li-Chan, E.C.Y. (2014) Investigation of the
putative associations between dairy consumption and incidence
of type 1 and type 2 diabetes. Critical Reviews in Food Science
and Nutrition, 54, 411432.
20 Knekt, P., Laaksonen, M., Mattila, C. et al. (2008) Serum vitamin D and subsequent occurrence of type 2 diabetes. Epidemiology, 19, 666671.
21 Pittas, A.G., Sun, Q., Manson, J.E. et al. (2010) Plasma 25hydroxyvitamin D concentration and risk of incident type 2
diabetes in women. Diabetes Care, 33, 20212023.
22 Anderson, J.L., May, H.T., Horne, B.D. et al. (2010) Relation of
vitamin D deficiency to cardiovascular risk factors, disease status,
and incident events in a general healthcare population. The
American Journal of Cardiology, 106, 963968.
23 Liu, E., Meigs, J.B., Pittas, A.G. et al. (2010) Predicted 25hydroxyvitamin D score and incident type 2 diabetes in the
Framingham Offspring Study. American Journal of Clinical Nutrition, 91, 16271633.
24 Mitri, J., Muraru, M.D. & Pittas, A.G. (2011) Vitamin D and
type 2 diabetes: a systematic review. European Journal of Clinical
Nutrition, 65, 10051015.
25 de Boer, I.H., Tinker, L.F., Connelly, S. et al. (2008) Calcium
plus vitamin D supplementation and the risk of incident diabetes
in the Womens Health Initiative. Diabetes Care, 31, 701707.
26 Avenell, A., Cook, J.A., MacLennan, G.S. et al. (2009) Vitamin D
supplementation and type 2 diabetes: a substudy of a randomised placebo-controlled trial in older people (RECORD trial,
ISRCTN 51647438). Age and Ageing, 38, 606609.
27 Muniyappa, R., Lee, S., Chen, H. et al. (2008) Current
approaches for assessing insulin sensitivity and resistance in vivo:
advantages, limitations, and appropriate usage. American Journal
of Physiology Endocrinology & Metabolism, 294, E15E26.
28 Forouhi, N.G., Luan, J., Cooper, A. et al. (2008) Baseline serum
25-hydroxy vitamin d is predictive of future glycemic status and
insulin resistance: the Medical Research Council Ely Prospective
Study 1990-2000. Diabetes, 57, 26192625.
29 Muscogiuri, G., Sorice, G.P., Prioletta, A. et al. (2010) 25Hydroxyvitamin D concentration correlates with insulinsensitivity and BMI in obesity. Obesity, 18, 19061910.
30 Rajakumar, K., de las Heras, J., Lee, S. et al. (2012) 25-hydroxyvitamin D concentrations and in vivo insulin sensitivity and
beta-cell function relative to insulin sensitivity in black and
white youth. Diabetes Care, 35, 627633.
31 de las Heras, J., Rajakumar, K., Lee, S. et al. (2013) 25-Hydroxyvitamin D in obese youth across the spectrum of glucose tolerance from normal to prediabetes to type 2 diabetes. Diabetes
Care, 36, 20482053.

Clinical Endocrinology (2016) 84, 159171

doi: 10.1111/cen.12760

REVIEW ARTICLE

Vitamin D and insulin resistance

Ian R. Wallace*,, Helen J. Wallace*,, Michelle C. McKinley, Patrick M. Bell* and Steven J. Hunter*
*Regional Centre for Endocrinology and Diabetes, Royal Victoria Hospital and Nutrition and Metabolism Group, Centre for Public
Health, Queens University Belfast, Belfast, UK

Introduction
Summary
Vitamin D is a steroid hormone, which in active form binds to
the vitamin D receptor. Expression of the vitamin D receptor in
diverse cell types (pancreatic islet cells, myocytes, hepatocytes
and adipocytes) raises the suspicion that vitamin D may be
involved in multiple cellular processes, including the response to
insulin. Insulin resistance is a characteristic feature of type 2
DM, and its attenuation may reduce the incidence of type 2 DM
and cardiovascular disease. In observational studies, low serum
25-hydroxyvitamin D (25-OHD) concentrations are associated
with an increased risk of type 2 DM. It has been suggested that
increasing serum 25-OHD concentrations may have beneficial
effects on glucose and insulin homeostasis. However, cross-sectional and interventional studies of vitamin D supplementation
provide conflicting results and demonstrate no clear beneficial
effect of vitamin D on insulin resistance. These studies are
complicated by inclusion of different patient cohorts, different
25-OHD assays and different doses and preparations of vitamin
D. Any possible association may be confounded by alterations in
PTH, 1,25-dihydroxyvitamin D or tissue vitamin D concentrations. We identified 39 studies via MEDLINE and PUBMED.
We review the evidence from 10 studies (seven observational
and three interventional) examining vitamin D and type 2 DM
incidence, and 29 studies (one prospective observational, 12
cross-sectional and 16 interventional trials) examining vitamin
D and insulin resistance. Based on this data, it is not possible to
state that vitamin D supplementation has any effect on type 2
DM incidence or on insulin resistance. Data from the multiple
ongoing randomized controlled trials of vitamin D supplementation due to report over the next few years should help to clarify
this area.
(Received 20 October 2014; returned for revision 6 November
2014; finally revised 24 February 2015; accepted 26 February 2015)

Correspondence: Dr Ian Wallace, Regional Centre for Endocrinology and

Diabetes, Royal Victoria Hospital, Grosvenor Road, Belfast BT12 6BA,
UK. Tel.: 028 9063 4462; Fax: 028 9031 0111;
E-mail: iwallace01@qub.ac.uk
2015 John Wiley & Sons Ltd

Vitamin D is a steroid hormone, which in active form binds to

the vitamin D receptor. Expression of the vitamin D receptor in
diverse cell types throughout the body raises the suspicion that
vitamin D may be involved in multiple cellular processes,
including the response to insulin. Insulin resistance is defined as
a reduced biological response to insulin.1 It is strongly associated
with cardiovascular disease (CVD), is a characteristic feature of
type 2 DM,2,3 and its attenuation may reduce the incidence of
type 2 DM and CVD. In this article, we describe the physiology
of vitamin D and review evidence of an association between
serum 25-hydroxyvitamin D (25-OHD) concentration and insulin resistance.

Vitamin D physiology and mode of action

The term vitamin D refers to a group of sterols including
vitamin D2 (ergocalciferol) and vitamin D3 (cholecalciferol).
Vitamin D may be synthesized by skin cells in response to ultraviolet irradiation (UVB) or obtained from dietary sources. Plant
sources contain vitamin D2 whereas sunlight and animal sources
provide vitamin D3. UVB irradiation of the precursor molecule,
7-dehydrocholesterol, leads to formation of previtamin D. Vitamin D undergoes two hydroxylation steps, first in the liver and
subsequently in the proximal convoluted tubule of the kidney,
resulting in the active form (1,25-dihydroxyvitamin D).
The human vitamin D receptor is a 50- to 60-kDa molecule
with strong affinity for the active 1,25-dihydroxyvitamin D form.
The vitamin D receptor is present in diverse cell types. Pancreatic islet cells have both vitamin D receptors and vitamin
D-dependent calcium-binding proteins. Insulin secretion is
dependent on changes in intracellular calcium concentration,
and 1,25-dihydroxyvitamin D has been shown to regulate betacell calcium flux in vitro and in mouse models.4 Likewise, the
intracellular response to insulin can also be affected by alterations in intracellular calcium concentration.5 The vitamin D
receptor is also highly expressed in adipocytes. Activation modulates intracellular calcium concentrations and can effect changes
in lipogenesis and lipolysis.4 It is therefore plausible that vitamin
D may be involved in beta-cell secretory activity and in modulating the tissue response to insulin. It is not yet clear whether
the vitamin D receptor is expressed in myocytes and vitamin D
effects on muscle insulin sensitivity, if any, may be indirect.6,7
159

Vitamin D and insulin resistance 171

64

65

66

67

meta-analysis of observational cohort and randomised intervention studies. BMJ, 348, g1903.
Neyestani, T.R., Djazayery, A., Shab-Bidar, S. et al. (2013) Vitamin D Receptor Fok-I polymorphism modulates diabetic host
response to vitamin D intake: need for a nutrigenetic approach.
Diabetes Care, 36, 550556.
OSullivan, A., Gibney, M.J., Connor, A.O. et al. (2011) Biochemical and metabolomic phenotyping in the identification of a
vitamin D responsive metabotype for markers of the metabolic
syndrome. Molecular Nutrition & Food Research, 55, 679690.
Autier, P., Boniol, M., Pizot, C. et al. (2014) Vitamin D status
and ill health: a systematic review. The Lancet Diabetes and
Endocrinology, 2, 7689.
Ashraf, A.P., Huisingh, C., Alvarez, J.A. et al. (2014) Insulin
resistance indices are inversely associated with vitamin D binding

2015 John Wiley & Sons Ltd

Clinical Endocrinology (2016), 84, 159171

protein concentrations. Journal of Clinical Endocrinology &

Metabolism, 99, 178183.
68 Pepe, J., Romagnoli, E., Nofroni, I. et al. (2005) Vitamin D status as the major factor determining the circulating levels of parathyroid hormone: a study in normal subjects. Osteoporosis
International, 16, 805812.
69 Seida, J., Mitri, J., Colmers, I. et al. (2014) Effect of vitamin
D3 supplementation on improving glucose homeostasis and
preventing diabetes: a systematic review and meta-analysis.
Journal of Clinical Endocrinology & Metabolism, 99, 3551
3560.
70 Bolland, M.J., Grey, A., Gamble, G.D. et al. (2014) The effect of
vitamin D supplementation on skeletal, vascular, or cancer outcomes: a trial sequential meta-analysis. The Lancet Diabetes and
Endocrinology, 2, 307320.