Does honey have any salutary effect against streptozotocin - induced diabetes in rats?
© The Author(s). 2016
Received: 9 May 2016
Accepted: 29 November 2016
Published: 24 January 2017
Diabetes is a global, growing and costly public health problem. In the literature, there are conflicting reports on the effect of consumption of bee honey on diabetes. We assessed the possible effect of a commercially available bee honey (given orally by gavage at doses of 1 g/kg/day for 4 weeks) on the blood concentrations of glucose, insulin and leptin and body weight of rats with streptozotocin-induced diabetes.
Thirty-six rats were allocated randomly into six groups equally and treated for 4 weeks as follows: Group.1: non-diabetic rats given distilled water, group.2: non-diabetic rats given honey (1 g/kg), group.3: Diabetic rats given distilled water, group.4: Diabetic rats given honey, group.5: Diabetic rats given insulin (10 IU/kg), and group.6: Diabetic rats given combination of insulin (10 IU/kg) with honey (1 g/kg). The body weight, blood glucose, insulin and leptin concentrations of each rat were measured.
Honey treatment did not significantly affect the glucose, leptin and insulin concentrations of diabetic rats. It did not significantly affect the excessive water intake or urinary output in diabetic rats when compared to the insulin-treated groups. Neither honey nor insulin improved body weight in diabetic rats.
Contrary to the reports of a salutary effect of honey in diabetic humans and rodents, our results showed that consumption of honey caused no significant changes in body weight, or glucose and insulin concentrations. However, further studies with different doses and durations of treatment are warranted.
KeywordsDiabetes Honey Insulin Blood glucose Body weight Leptin
Diabetes is an abnormal metabolic state that results in hyperglycemia due to either insulin deficiency or abnormal response to insulin, and is of two types: type I, which is caused by severe deficiency in insulin synthesis, and type II diabetes that is complicated with insulin resistance and/or insulin production deficiency . In 2014, it has been estimated that there are about 387 million diabetic patients worldwide . Consequently, studies conducted to find out an ideal remedy against diabetes are needed, and bee honey has been suggested as one of these [3, 4].
Bees honey from bees (Apis mellifera) is produced from nectar and it consists of at least 181components including fructose (37.5%), glucose (30.6%), sucrose (1.62%), maltose (2.7%), water (17.2%), vitamins, minerals, enzymes, acids such as flavonoids, phenolic acids and other components [3–9]. Honey’s constituents differ slightly depending on the botanical species [10, 11].
In folk medicine, honey is used in different parts of the world for a number of unrelated diseases and conditions . Furthermore, it is known to possess antioxidant, anti-inflammatory and antimicrobial properties [4, 11–13]. Moreover, it has been reported that honey has effects against hyperglycemia in diabetic patients and against complications of diabetes. Its anti-hyperglycemic effects may relate to its fructose content and antioxidant properties [4, 6]. Fructose is a monosaccharide sugar that has slow absorption rate and its intake may lead to delaying the digestion and elongating the gastric emptying. Therefore, the fructose was suggested to contribute to the anti-hyperglycemic effects of the honey. Some studies showed that fructose increases hepatic uptake of glucose resulting in decreasing blood glucose . The antioxidant effects of honey may be due to compounds such as flavonoids and phenolic acids [5, 14–16]. These antioxidants activities of honey may improve oxidative stress in β-cells in the pancreas, and this may lead to promoting insulin secretion. In addition, honey has been found to reduce insulin resistance in type II diabetic patients .
In literature, many parameters were used to investigate glycemic control of diabetic patients/animals including blood glucose, body weight, HbA1c and others. Moreover, many other parameters (eg insulin and leptin) were suggested to be used as indicators or predictors for diabetes status or its glycemic control. There is a controversy regarding the association between leptin levels & diabetes. However, it has been reported that blood leptin levels may be used as a predictor for glycemic control . Moreover, Japanese researchers investigated the possible therapeutic effects of leptin in congenital hyperleptinemic diabetic mice models where it showed a salutary effect . Many studies were conducted to find out the effect of honey on diabetic rats in which they investigated blood insulin level and their results were variable [20–22].
In literature, there is a controversy regarding the antihyperglycemic effects of honey on diabetes. Therefore, our study aimed to assess the effect of a commercially available bee honey (1 g/kg/day for 4 weeks) on the concentrations of blood glucose, insulin and leptin and body weight in normal rats and rats with streptozotocin-induced diabetes.
This prospective randomized controlled study was carried at Sultan Qaboos University in Muscat, Oman between January 2013 and July 2014. The use of rats and all the experimental procedures were approved by the Animal Ethical Committee at Sultan Qaboos University (SQU/ACE/2012-13/4). Thirty six female Sprague Dawley rats (250–350 g) were obtained from Small Animal House of Sultan Qaboos University, Muscat, Oman and were kept under standard conditions (a temperature of 22 ± 2 °C, relative humidity of about 60%, with a 12 h light–dark cycle (lights on 6:00), and (unless otherwise mentioned) given free access to standard pellet chow diet containing 0.85% phosphorus, 1.12% calcium, 0.35% magnesium, 25.3% crude protein and 2.5 IU/g vitamin D3 (Oman Flour Mills, Muscat, Oman) and water. Ethical clearance was obtained from our University Animal Ethics Committee and all procedures involving animals and their care were carried out in accordance with international laws and policies (EEC Council directives 86/609, OJL 358, 1 December, 12, 1987; NIH Guide for the Care and Use of Laboratory Animals, NIH Publications No. 85-23, 1985).
Induction of diabetes
Diabetes was induced after 18 h fasting by intraperitoneal (i.p.) injection of streptozotocin (STZ) at a dose of 50 mg/kg, dissolved in 0.1 M of a citrate buffer (pH 4.5) . Other groups of rats were injected with citrate buffer. Two days thereafter, the blood glucose (BG) concentration was checked using a OneTouch® UltraMini® Meter (LifeScan, Milpitas, CA, USA) using a drop of blood obtained by pricking tip tail. Rats with blood glucose concentrations ≥ 14 mmol/L were considered diabetic . Treatments were started 3 weeks after STZ or citrate buffer injections. Body weight was measured daily, fasting blood glucose concentration was measured weekly, and blood insulin & leptin concentrations were measured once at the end of the experiment.
The animals were allocated randomly into six equal groups and treated for 4 weeks as follows: Group1 (C): non-diabetic and were treated via oral gavage with distilled water (0.5 ml/rat) via oral gavage once daily; Group 2 (H): non-diabetic and were given honey (1 g/kg body weight) via oral gavage once daily [12, 20, 24]; Group 3 (D): diabetic and were treated with distilled water (0.5 ml/rat) via oral gavage once daily; Group 4 (DH): diabetic and were treated with honey (1 g/kg body weight) via oral gavage once daily; Group 5 (DI): diabetic and were treated with insulin (10 IU/Kg body weight) subcutaneously once daily ; and Group 6 (DIH): diabetic and were treated with honey (1 g/kg body weight) and insulin (10 IU/Kg body weight) once daily.
After 4 weeks of treatment, rats were placed in metabolic cages to measure water intake & collect 24-h urine. Fasting Blood glucose concentration was measured and rats were then anesthetized (using ketamine (75 mg/kg) and xylazine (5 mg/kg) intraperitoneally) and blood (4 ml) collected from the inferior vena cava. Blood samples were centrifuged (900 g for 15 min at 4 °C) and plasma collected was frozen for subsequent measurement of leptin and insulin concentrations.
Commercially available honey (Golden Glory, Australia) was bought from a local market in Muscat. Honey (1 g/kg/day) was diluted with distilled water before given it to the rats by gavage. As well as, Insulin (10 IU/Kg body weight) was diluted with 0.90% NaCl just before injecting.
Collected plasma samples were analyzed using ELISA test for both insulin (EIA Kit, Gayman Chemical Company, Michigan, USA) and leptin (EIA Kit, Gayman Chemical Company, Michigan, USA).
Statistical analysis was conducted by the GraphPad Prism 6.0 (GraphPad Software, San Diego, CA, USA). Each group consisted of six rats. All data are given as means ± S.E.M. Group means were compared with an analysis of variance (ANOVA; One-Way) followed by Tukey’s multiple comparison test. Values of P < 0.05 were considered significant.
Urine output of the diabetic control group was significantly (p < 0.05) higher than non-diabetic control. The diabetic group that treated with honey alone showed significant increase in urine output compared to non-diabetic control (p < 0.0001) and non-diabetic that received honey (p < 0.001). Moreover, diabetic with insulin and diabetic that treated with insulin in combination with honey showed significant decrease in urine output compared to diabetic with honey (p < 0.01).
The effect of honey on body weight of normal and diabetic rats
Diabetic + Honey
Diabetic + Insulin
Diabetic + Insulin + Honey
A week before treatment
302.00 ± 4.95
241.17 ± 12.08 ** ##
303.00 ± 13.92
233.17 ± 12.32 ** # #
242.00 ± 10.61* #
261.60 ± 6.86
Weeks after treatment
307.50 ± 5.80
230.00 ± 13.68*** ##
300.60 ± 13.71
220.00 ± 12.25**** ###
241.25 ± 16.18** #
252.20 ± 6.15*
328.50 ± 6.16
260.00 ± 14.76** #
322.00 ± 15.43
238.17 ± 14.43*** ##
252.25 ± 16.62** #
251.20 ± 5.44** ##
327.83 ± 6.80
239.17 ± 14.08**** ###
323.60 ± 13.82
224.67 ± 8.83**** ####
253.50 ± 16.81** ##
251.00 ± 7.12** ##
324.83 ± 6.16
202.33 ± 14.44 *** ####
324.00 ± 12.35
182.33 ± 14.37 *** ###
256.25 ± 17.72 * # ΨΨ
250.60 ± 6.59 ** ## ΨΨ
The effect of honey on blood glucose concentration in the blood of normal and diabetic rats
Diabetic + Honey
Diabetic + Insulin
Diabetic + Insulin + Honey
A week before treatment
4.25 ± 0.25
13.07 ± 2.48
4.98 ± 0.22
19.60 ± 3.57 ** ##
17.60 ± 3.25 * #
25.74 ± 2.28 **** $ ####
Weeks after treatment
5.13 ± 0.24
23.85 ± 0.83 **** ###
5.00 ± 0.22
22.72 ± 3.90 *** ###
11.65 ± 2.43 $
17.30 ± 4.00 * #
4.02 ± 0.10
14.20 ± 3.55
4.16 ± 0.09
15.08 ± 2.86
5.95 ± 0.85
13.80 ± 4.52
4.00 ± 0.22
19.05 ± 3.52 ** ##
3.88 ± 0.11
26.37 ± 2.35 **** ####
13.78 ± 4.46 Ψ
15.40 ± 3.59
4.82 ± 0.21
21.55 ± 2.97 *** ##
4.90 ± 0.16
22.52 ± 3.72 *** ###
8.33 ± 2.15 $, Ψ
11.48 ± 2.77 Ψ
The diabetic control group had significantly (p < 0.01) lower blood leptin concentration compared to non-diabetic control rats and non-diabetic group that received honey alone. There is no significant difference between non-diabetic groups and diabetic groups that received honey, insulin or their combination (ie honey improved leptin blood concentration).
In the current study, we found that honey treatment did not significantly prevent excessive water intake and urinary frequency in diabetic rats. However, in rats treated with insulin, there were improvements in these two physiological parameters. As far as we are aware, there are no previously reported data on the effects of honey and insulin (given together) on water intake and urine output.
As expected, significant decrease in body weight was observed in STZ-induced diabetic rats, and this may be because of connective tissue degradation due to the surge lysosomal activity . Our study showed that neither honey nor insulin improved the loss in body weight induced by diabetes when compared to non-diabetic rats. In contrast, some previous studies had reported that honey has a beneficial effect in increasing the body weight of the diabetic humans and rats probably due to the fructose component of honey with its purported antihyperglycemic effects [12, 20]. Diabetic rats that were treated with insulin showed body weight gain and this may be due to the effect of insulin in decreasing energy loss via urine .
Unlike our present results, some authors in Malaysia [12, 20] reported that the honey they used in their experiments was effective in ameliorating streptozotocin - induced diabetes in rats because honey has anti-oxidant properties against streptozotocin – induced oxidative stress, which leads to β cells destruction. Fructose in the honey may also slows the absorption rate which may lead to delaying the digestion and elongating the gastric emptying and its ability to increase hepatic uptake of glucose resulting in decreasing blood glucose . The antioxidant effects of honey may be attributed to its flavonoids and phenolic acids contents, which decrease the free radicals in β-cells in the pancreas, and this may lead to promoting insulin secretion [5, 14–16]. However, Bahrami and his colleagues reported that honey had no significant difference in blood glucose concentration of diabetic patients with type 2 diabetes. When honey used to treat diabetic rats, they showed no improvement in blood glucose concentration .
It is not clear why we could not reproduce the positive results of those who claim honey is beneficial in diabetic rats [12, 20]. The discrepancy may be due to the type of honey used in both experiments. It is known that the composition of honey depends on its floral origin, or to other unknown factors [10, 11].
Plasma insulin concentration is usually significantly reduced in rats with streptozotocin – induced diabetes . However, we found that there is no significant difference in blood insulin concentration among all groups except for diabetic rats given honey in combination with insulin and non-diabetic group given honey. Many studies were conducted to find out the effect of honey on diabetic rats and their results were at variance. Some authors found that honey increases the concentration of insulin concentration but the others showed that honey had no effect on blood insulin concentration [20–22]. Our results seem to insignificantly inceases the plasma concnetrations of insulin.
Leptin is a protein hormone mainly produced by adipocytes and plays a vital role in the regulation of energy intake and expenditure, and is positively associated with body mass index (BMI) and the absolute fat mass, as well as glycemic control . The recent study showed that honey increased leptin blood concentration of diabetic rats given honey, insulin or their combination compared to diabetic control group. Many studies were conducted to investigate the association between leptin and diabetes and their results were contradicted. Some studies showed that there is a direct association between leptin and diabetes development [30–33]. On the other hand, other studies found that there is no association between leptin and diabetes [34, 35]. A randomized control trial on rats showed that the rats treated with honey have lower blood leptin concentration compared to those treated with sucrose .
The possible limitations of this experiment are the small number of rats in each group and the use of a single type of honey at a single dose. Further experiments using different types of honey at different doses against streptozotocin-induced diabetes are warranted.
Based on our results, the present preliminary experiment suggests that the use of honey as a possible treatment in streptozotocin-induced diabetes in rats is not beneficial.
We thank the technical staff of the Department of Pharmacology, COMHS & Small Animal House of SQU, for their help.
The work was supported by College of Medicine & Health Sciences, Sultan Qaboos University, Oman.
Availability of data and materials
Raw data of this work is kept in our database, and will be made available when required.
ZMA participated in the design of the study, collected data and wrote draft of the manuscript. BHA designed the study, assisted in data collection and manuscripts preparation & writing. Both authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
The use of rats and all the experimental procedures were approved by the Animal Ethical Committee at Sultan Qaboos University (SQU/ACE/2012-13/4).
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- American Diabetes Association. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2012;36(Suppl l):S67–74.PubMed CentralGoogle Scholar
- International Diabetes Federation: Diabetes Atlas. 2014. http://www.idf.org/diabetesatlas/5e/mortality. Accessed 15 Oct 2015.
- Erejuwa O. Effect of honey in diabetes mellitus: matters arising. J Diabetes Metab Disord. 2014;13(1):23.View ArticlePubMedPubMed CentralGoogle Scholar
- Erejuwa O, Sulaiman S, Wahab M. Honey - a novel antidiabetic agent. Int J Biol Sci. 2012;8(6):913–34.View ArticlePubMedPubMed CentralGoogle Scholar
- Ediriweera E, Premarathna N. Medicinal and cosmetic uses of Bee’s honey - a review. AYU. 2012;33(2):178–82.View ArticlePubMedPubMed CentralGoogle Scholar
- Mohamed M, Sulaiman S, Jaafar H, Sirajudeen K. Effect of different doses of Malaysian honey on reproductive parameters in adult male rats. Andrologia. 2011;44:182–6.View ArticlePubMedGoogle Scholar
- Sajwani A, Eltayeb E, Farook S, Patzelt A. Sugar and Protein Profiles of Omani Honey from Muscat and Batinah Regions of Oman. Int J Food Prop. 2007;10(4):675–90.View ArticleGoogle Scholar
- Qiu P, Ding H, Tang Y, Xu R. Determination of chemical composition of commercial honey by near-infrared spectroscopy. J Agric Food Chem. 1999;47(7):2760–5.View ArticlePubMedGoogle Scholar
- Abdulrhman M, El Hefnawy M, Ali R, Abdel Hamid I, Abou El-Goud A, Refai D. Effects of honey, sucrose and glucose on blood glucose and C-peptide in patients with type 1 diabetes mellitus. Complement Ther Clin Pract. 2013;19(1):15–9.View ArticlePubMedGoogle Scholar
- Bahrami M, Ataie-Jafari A, Hosseini S, Foruzanfar M, Rahmani M, Pajouhi M. Effects of natural honey consumption in diabetic patients: an 8-week randomized clinical trial. Int J Food Sci Nutr. 2009;60(7):618–26.View ArticlePubMedGoogle Scholar
- Al-Jabri A, Al-Hosni S, Nzeako B, Al-Mahrooqi Z, Nsanze H. Antibacterial activity of Omani honey alone and in combination with gentamicin. Saudi Med J. 2005;26(5):767–71.PubMedGoogle Scholar
- Erejuwa O, Sulaiman S, Wahab M, Sirajudeen K, Salleh M, Gurtu S. Antioxidant Protective Effect of Glibenclamide and Metformin in Combination with Honey in Pancreas of Streptozotocin-Induced Diabetic Rats. Int J Mol Sci. 2010;11(5):2056–66.View ArticlePubMedPubMed CentralGoogle Scholar
- Moghazy A, Shams M, Adly O, Abbas A, El-Badawy M, Elsakka D, et al. The clinical and cost effectiveness of bee honey dressing in the treatment of diabetic foot ulcers. Diabetes Res Clin Pract. 2010;89(3):276–81.View ArticlePubMedGoogle Scholar
- Zaid S, Sulaiman S, Sirajudeen K, Othman N. The effects of tualang honey on female reproductive organs, tibia bone and hormonal profile in ovariectomised rats - animal model for menopause. BMC Complement Altern Med. 2010;10(1):82.View ArticlePubMedPubMed CentralGoogle Scholar
- Batumalaie K, Qvist R, Yusof K, Ismail I, Sekaran S. The antioxidant effect of the Malaysian Gelam honey on pancreatic hamster cells cultured under hyperglycemic conditions. Clin Exp Med. 2014;14(2):185–95.View ArticlePubMedGoogle Scholar
- Al-Waili N, Salom K, Al-Ghamdi A, Ansari M, Al-Waili A, Al-Waili T. Honey and Cardiovascular Risk Factors, in Normal Individuals and in Patients with Diabetes Mellitus or Dyslipidemia. J Med Food. 2013;16(12):1063–78.View ArticlePubMedGoogle Scholar
- Katsilambros N, Philippides P, Touliatou A, Georgakopoulos K, Kofotzouli L, Frangaki D, et al. Metabolic effects of honey (alone or combined with other foods) in type II diabetics. Acta Diabetol Lat. 1988;25(3):197–203.View ArticlePubMedGoogle Scholar
- Zhou L, Cai X, Zhu Y, Liu W, Gong S, Zhang S, et al. Serum leptin level is associated with glycaemic control in newly diagnosed type 2 diabetes patients: A 1-year cohort study. Diabetes Metab. 2016.
- Naito M, Fujikura J, Ebihara K, Miyanaga F, Yokoi H, Kusakabe T, et al. Therapeutic impact of leptin on diabetes, diabetic complications, and longevity in insulin-deficient diabetic mice. Diabetes. 2011;60(9):2265–73.View ArticlePubMedPubMed CentralGoogle Scholar
- Erejuwa O, Sulaiman S, Wahab M, Sirajudeen K, Salleh M, Gurtu S. Glibenclamide or metformin combined with honey improves glycemic control in streptozotocin-induced diabetic rats. Int J Biol Sci. 2011;7(2):244–52.View ArticlePubMedPubMed CentralGoogle Scholar
- Nasrolahi O, Khaneshi F, Rahmani F, Razi M. Honey and metformin ameliorated diabetes-induced damages in testes of rat; correlation with hormonal changes. Iran J Reprod Med. 2013;11(12):1013–20.PubMedPubMed CentralGoogle Scholar
- Nemoseck T, Carmody E, Furchner-Evanson A, Gleason M, Li A, Potter H, et al. Honey promotes lower weight gain, adiposity, and triglycerides than sucrose in rats. Nutr Res. 2011;31(1):55–60.View ArticlePubMedGoogle Scholar
- Wójcicka G, Jamroz-Wiśniewska A, Marciniak A, Łowicka E, Bełtowski J. The differentiating effect of glimepiride and glibenclamide on paraoxonase 1 and platelet-activating factor acetylohydrolase activity. Life Sci. 2010;87(3–4):126–32.View ArticlePubMedGoogle Scholar
- Erejuwa O, Sulaiman S, Wahab M, Sirajudeen K, Salleh M, Gurtu S. Differential responses to blood pressure and oxidative stress in streptozotocin-induced diabetic Wistar-Kyoto rats and spontaneously hypertensive rats: effects of antioxidant (Honey) treatment. Int J Mol Sci. 2011;12(3):1888–907.View ArticlePubMedPubMed CentralGoogle Scholar
- Elvert R, Wille A, Wandschneider J, Werner U, Glombik H, Herling A. Energy loss via urine and faeces - a combustive analysis in diabetic rats and the impact of antidiabetic treatment on body weight. Diabetes Obes Metab. 2013;15(4):324–34.View ArticlePubMedGoogle Scholar
- Rajkumar L, Srinivasan N, Balasubramanian K, Govindarajulu P. Increased degradation of dermal collagen in diabetic rats. Indian J Exp Biol. 1991;29(11):1081–3.PubMedGoogle Scholar
- Abdul Sani N, Belani L, Pui Sin C, Abdul Rahman S, Das S, Zar Chi T, et al. Effect of the combination of gelam honey and ginger on oxidative stress and metabolic profile in streptozotocin-induced diabetic Sprague-Dawley rats. Biomed Res Int. 2014;2014:1–9.View ArticleGoogle Scholar
- Meena S, Rajput YS, Pandey AK, Sharma R, Singh R. Camel milk ameliorates hyperglycaemia and oxidative damage in type-1 diabetic experimental rats. J Dairy Res. 2016;83:412–9.View ArticlePubMedGoogle Scholar
- Meek TH, Morton GJ. The role of leptin in diabetes: metabolic effects. Diabetologia. 2016;59:928–32.View ArticlePubMedGoogle Scholar
- Wannamethee S, Lowe G, Rumley A, Cherry L, Whincup P, Sattar N. Adipokines and risk of type 2 diabetes in older men. Diabetes Care. 2007;30(5):1200–5.View ArticlePubMedGoogle Scholar
- Schmidt M, Duncan B, Vigo A, Pankow J, Couper D, Ballantyne C, et al. Leptin and incident type 2 diabetes: risk or protection? Diabetologia. 2006;49(9):2086–96.View ArticlePubMedGoogle Scholar
- Franks P, Brage S, Luan J, Ekelund U, Rahman M, Farooqi I, et al. Leptin predicts a worsening of the features of the metabolic syndrome independently of obesity. Obes Res. 2005;13(8):1476–84.View ArticlePubMedGoogle Scholar
- McNeely M, Boyko E, Weigle D, Shofer J, Chessler S, Leonnetti D, et al. Association between baseline plasma leptin levels and subsequent development of diabetes in Japanese Americans. Diabetes Care. 1999;22(1):65–70.View ArticlePubMedGoogle Scholar
- Vinitha R, Ram J, Snehalatha C, Nanditha A, Shetty A, Arun R, et al. Adiponectin, leptin, interleukin-6 and HbA1c in the prediction of incident type 2 diabetes: A nested case-control study in Asian Indian men with impaired glucose tolerance. Diabetes Res Clin Pract. 2015;109(2):340–6.View ArticlePubMedGoogle Scholar
- Kanaya A, Wassel Fyr C, Vittinghoff E, Harris T, Park S, Goodpaster B, et al. Adipocytokines and incident diabetes mellitus in older adults. Arch Intern Med. 2006;166(3):350–6.View ArticlePubMedGoogle Scholar