Musaiger and Abdulaziz (1986) studied the demographic characteristics of diabetic patients in Bahrain. The study included all patients with diabetes as the first diagnosis between the years 1980 and 1982. About 88% of the diabetics were found to be native Bahrainis. The highest number of diabetics were found in the over 50-years age group. In a worrying trend, however, the number of diabetics under 10 years of age was seen to increase dramatically from 0.8% in 1980 to 11.1% in 1982. Musaiger and Abdulaziz reckoned that this could be due to the pancreatic infection by viruses such as mumps or influenza that are common in Bahrain. Female diabetics were found to be much higher than the males among all age groups, except in the under 10-years group. The authors attributed the high incidence of diabetes in Bahraini women to obesity, inactivity, and diet. Similarly, the prevalence of diabetes was found to be very high in urban as compared to rural areas. Muharraq city and Manama reported the highest number of diabetes cases, whereas the Western Region, Sitra, and the Central Area had very few cases. This again was attributed to the lifestyle in cities with inactivity, obesity, and consumption of high energy foods. Musaiger and Abdulaziz (1986) recommended that health workers should be trained on diet management for diabetics. [Musaiger AO, Abdulaziz SA. Demographic characteristics of hospitalised patients with diabetes in Bahrain. Bahrain Med Bull. 1986; 8(2):73-6.]

Zurba and Al-Garf (1996) undertook an epidemiological study to estimate the prevalence of diabetes mellitus among Bahraini nationals above 20-years of age. A total of 572 individuals attending four health centers in Bahrain for any problem were randomly selected as the study group. Data pertaining to the subjects' age, sex, personal and family history of diabetes, hypertension, weight, and height, were collected. In addition, the blood glucose levels were estimated 2-hours post-75 g oral glucose. Subjects known to have been diagnosed with diabetes were exempted from the blood test. The mean age of the study group was 43.9 years, and the male to female ratio was 1:1.9. Interestingly, 27.6% of the total subjects studied had a positive family history of diabetes, whereas 41.7% of the subjects diagnosed with diabetes had a positive family history. Total prevalence of diabetes among this study group was found to be 25.5% (males-26.4%, females-25%). Prevalence of known diabetes (17.3%) was higher than that of newly diagnosed (8.2%) cases. Zurba and Al-Garf (1996) surmised that this could be indicative of the increased awareness among the population, as well as the easy accessibility of free health care services. About 14.7% of the subjects were found to have impaired glucose tolerance (IGT). In addition, 58.3% of the subjects diagnosed with diabetes and 53.4% of the subjects with IGT showed increased blood pressure according to the WHO expert group criteria for hypertension (1994). Almost 75% of subjects with diabetes and a similar percentage of subjects with IGT were found to be obese (BMI > 25%). Gross obesity (BMI > 30%) was observed in 31.5% of subjects with diabetes and 34.2% in cases with IGT. [Zurba FI, Al-Garf A, Prevalence of diabetes mellitus in Bahrain. Bahrain Med Bull. 1996; 18(2):44-51.]

Al-Mahroos and McKeigue (1998) performed the first cross sectional survey in Bahrain to detect the prevalence of diabetes mellitus among Bahraini natives. A total of 2128 Bahrainis, aged between 40 and 69-years were employed for the study. Parameters studied included the subjects' waist, hip and height measurements, family history of diabetes, physical activity, and blood glucose measurements after fasting as well as 2-hours post oral glucose administration. Based on the study, diabetes and IGT was estimated to be prevalent in 18% and 30% of the population, respectively. The highest numbers of diabetics were seen in the 55-59 years age group (31.9% males, 36.1% females). These prevalence rates were higher than those in most other populations compared, including other Arab populations from Kuwait, Saudi Arabia, and the UAE. There was a significant difference in the waist to hip ratio between diabetic (0.98 cm) and non-diabetic men (0.95 cm), whereas such a significant difference could not be seen among the women. The waist to height ratio however was significantly increased in both diabetic men and women as compared to their non-diabetic counterparts. The authors were able to show that only 6% of women and 20% of men between 50-59 years of age exercised by walking at least 1 Km each day. In the same age group, only 9% of men cycled, whereas only 7 women on the whole reported to be cycling. The results showed that about 35% of Bahraini diabetics between the ages of 40 and 69 years were undiagnosed. [Al-Mahroos F, McKeigue. Obesity, physical activity and prevalence of diabetes in Bahraini Arab native population. Bahrain Med Bull. 1998; 20(3):114-8.]

Bhatt and Samahiji (1999) performed a retrospective study of vitreous surgery to demonstrate the efficacy of the technique in the management of complications of diabetic retinopathy. Of all the surgeries, approximately 30% (22 eyes) were for complications of diabetic retinopathy. Of these 22 eyes, 16 belonged to females, and the remaining six to males. The diabetic patients ranged from age 32 years to 73 years. The surgeries were for vitreous hemorrhage alone (6), tractional retinal detachment with or without vitreous hemorrhage (13), or combined tractional and rhegomatogenous retinal detachment (3). Out of these, 20 (91%) of the eyes showed improved in visual acuity. Vision in two eyes did not improve, possibly due to optic atrophy and retinal ischemia. Three eyes required additional surgery, one underwent silicon oil removal, and one needed removal of recurrent vitreous hemorrhage. [Bhatt NS, Al-Samahiji S. Vitreous surgery for complications of diabetic retinopathy. Bahrain Med Bull. 1999; 21(4):113-7.]

In order to assess the effect of fasting during the holy month of Ramadan among diabetics, Al-Nasir (1999) studied 17 patients with NIDDM and four patients with both NIDDM and hypertension before, during, and after Ramadan. All patients showed a significant reduction in their weight 2-weeks (stage 1) and 4-weeks (stage 2) after the beginning of Ramadan (stage 0). However, this reduction was significant only in males. Fasting blood glucose was found to have significantly increased in stage 1 and reduced in stage 2. However, the pre- and post-fasting levels of FBG showed no statistically significant difference. Systolic and diastolic pressures decreased significantly in the period of study; more so the diastolic pressure. Al-Nasir (1999) was able to demonstrate that patients with NIDDM could observe fasting without any adverse effects, although they needed to be monitored closely during the period of fasting. [Al-Nasir FAL. Effect of Ramadan fasting on hypertension and diabetes. Qatar Med J. 1999; 8(2):53-7.]

Fikree et al. (2006) performed a retrospective study to determine the frequency of use of HbA1c level as an indicator for type 2 diabetes control and to identify the glycemic control among type 2 diabetic patients. A total of 383 (162 males and 221 females) type 2 diabetic patients were identified. The patients' ages ranged between 30 and 88 years. Oral treatment was taken by 327 patients, oral and insulin treatment by 24 cases, and insulin therapy by six cases. Six cases were on diet alone. One hundred seventy eight cases (46.5%) had HbA1c done for them and the mean HbA1c level was 9.6%. Among those, only 20 cases (11.2%) were controlled with an average HbA1c test result of 7% or less, while the rest (88.8%) were uncontrolled. There was no statistically significant relation between the HbA1c level and neither gender nor the type of treatment. [Fikree M, Hanafi B, Hussain ZA, Masuadi EM. Glycemic control of type 2 diabetes mellitus. Bahrain Med Bull. 2006; 28(3)]

Nemr et al. (2010) investigated the association of the c.C677T and c.A1298C mutations and diabetic nephropathy in patients with Type 2 DM. The study consisted of 252 Lebanese and 225 Bahraini subjects with Type 2 DM related nephropathy, and a control group of 309 Lebanese and 328 Bahraini patients with Type 2 DM in the absence of nephropathy. The Lebanese patients with nephropathy were found to have significantly higher frequencies of the 677T allele and C/C and T/T genotypes, when compared to Lebanese control patients. This did not hold for the Bahraini patients. The 1298C allele frequencies were comparable in the control and study groups. 677T/T carriers in both populations were found to have a higher homocysteine concentration. This increased concentration was linked with nephropathy in Lebanese, but not Bahraini patients. Regression analysis revealed the C677T mutation to be associated with nephropathy only among Lebanese patients. Nemr et al. (2010) stressed that the relationship between MTHFR mutations and diabetic nephropathy is dependent on the racial/ethnic background.

Al-Harbi et al. (2013) investigated the association between ACE polymorphisms and Type 2 Diabetes Mellitus in Bahrain. The case-control study involved 171 unrelated adult Bahraini T2DM patients and 188 age-matched Bahraini control subjects. PCR was used to detect two alleles: the 190 bp deletion (D allele) and 490 bp insertion (I allele). The frequencies of the II, ID, and DD genotypes were 13%, 39%, and 48%, respectively, in the T2DM patients, and 18%, 47%, and 35%, respectively, in the control subjects. DD genotype and D allele frequencies were significantly higher in the patients than in the controls. Al-Harbi et al. (2013) opined that the high level of D allele and DD genotype in the Bahraini population might be responsible for the high prevalence of T2DM in the country.

Mahmoud and Kamal (1998) studied 84 patients (57 males and 27 females; aged from 4 weeks to 58 years) with alopecia areata (AA). They found three patients to have diabetes mellitus and hypertension. Additionally, some patients were found to have a family history for diabetes mellitus and hypertension.

[Mahmoud SF, Kamal AM. Nail changes in minimal alopecia areata. Gulf J Dermatol. 1998; 5(1): 36-39.]

Mufunda et al. (2006) estimated the prevalence of diabetes to be ranging from < 1% in some rural areas to > 20% in some selected populations and racial groupings in urban settings. They also found that the predominant type of diabetes was type 2, which accounted for > 80% of all cases in some reports and tended to present later in life. Mufunda et al. (2006) found that recent studies based on analysis of hospital-based information have documented NCD trends that were similar to prevalence data generated from national risk factor surveys. Mufunda et al. (2006) concluded that NCD risk factors such as diabetes and hypertension are increasing in Africa.

Awadallah and Hamad (2003) conducted a study to investigate the association between haptoglobin (Hp) polymorphism and the occurrence of chronic renal failure (CRF) in Jordanians. The study group consisted of 159 patients with CRF resulting from various predisposing conditions and from 200 healthy unrelated controls. Hp phenotyping indicated that the Hp 2-2 phenotype was over-represented in CRF patients in general (0.547), patients with hypertension (0.622) and patients with diabetes mellitus (0.633). The Hp 2-1 phenotype was over-represented in patients with chronic glomerulonephritis (0.549) and patients with reflux nephropathy (0.445).

Taha et al. (1983) indicated that the annual incidence (1980-81) of diabetes mellitus among Kuwaitis under the age of 30-years is 22.09 per 100,000. The age distribution at onset showed an increase in incidence with age. The male/female sex ratio was 0.68, with the total number of female diabetics exceeding the males by 32%.

Richens et al. (1988) studied 75 Arab patients suffering from Type 2 Diabetes Mellitus. A family history of diabetes was present in 64% of the patients. Incidentally, patients with disease onset at under the age of 40-years had a significantly higher frequency of a positive family history, indicating a greater genetic influence in those with early onset.

Abdella (1990) detected a significant elevation in total alkaline phosphatase levels in 62 Arab diabetics in comparison to 64 healthy controls. Furthermore the bone isoenzyme fraction was significantly higher in diabetics. Five years later, Abdella et al. (1995) studied 3,299 patients (mean age-53 years; 1454 males and 1845 females) with Type 2 Diabetes Mellitus in order to examine any distinct clinical characteristics. Family history of diabetes in first-degree relatives was reported in 63% of the patients. The mean BMI was 31.8 in women and 28.5 in men. About 58% of the women and 34% of the men were obese, suggesting that obesity was a major risk factor for T2DM. In 1998, Abdella et al. carried out a separate study to determine prevalence rates of Type 2 DM in the Kuwaiti adult population aged over 20-years. In the study group of 30,003 subjects (1898 women), 14.8% were found to have NIDDM. Later, Abdella et al. (1998) also studied risk factors that could be linked to Type 2 DM and found that obesity, hypertension, and family history were significantly associated with the condition. They suggested that the aging Kuwaiti population would contribute to upward trends in the prevalence of abnormal glucose tolerance. [Abdella N. Serum alkaline phosphatase isoenzymes in an Arab population with non-insulin-dependent diabetes mellitus. Med Princ Pract. 1990; 2(3-4):215-20.]

Al-Sultan and Al-Zanki (2005) described the personal and clinical characteristics of type 2 diabetics in a health area in Kuwait. The study included 256 adult diabetics (55.5% men; mean age-51.7 years), 31.6% of whom were Kuwaitis. Mean BMI was found to be 32.6 among women and 27.8 among men. The most significant prevalent risk factor was smoking, while in women, it was presence of family history of diabetes, dieting, obesity, and hyperlipedemia. About 45% of the patients were obese, while 34% were overweight. High total cholesterol, triglycerides and LDL-cholesterol were found in 48%, 27%, and 72%, respectively. About 34% were diagnosed with metabolic syndrome. [Al-Sultan FA, Al-Zanki N. Clinical epidemiology of type 2 diabetes mellitus in Kuwait. Kuwait Med J. 2005; 37(2):98-104.]

Alattar et al. (2006) conducted a study to understand the prevalence of uncontrolled hypertension in Kuwaiti diabetics. They studied 251 diabetic patients, 85.3% of whom were suffering from Type 2 DM. Uncontrolled hypertension was detected in as high as 56.6% of the subjects. Statistical analysis led to the finding that factors such as old age, having type 2 DM rather than type 1 DM, longer duration of DM, obesity, past history of hypertension, proteinuria, and diabetic retinopathy were risk factors for hypertension among the diabetics. [Alattar AT, Alarouj M, Ben-Nakhi A, Hamadah A. Hypertension control among diabetes patients in Kuwait. Clin Diabetes. 2006; 5(1):42-6.]

Moussa et al. (2008) studied a randomly selected population of 128,918 Kuwaiti schoolchildren between the ages of 6 and 18-years to determine the prevalence of Type 2 DM in this population. A total of 45 of these children were found to have Type 2 DM (overall prevalence-34.9 per 100,000). Interestingly, males showed a significantly higher prevalence (46.3 vs 27.3). The prevalence also tended to increase with age. The age-adjusted prevalence of NIDDM in this population was calculated as 33.2 (41.6 males, 24.6 females). Affected children were also seen to have a significantly higher positive family history of the condition.

Barbari et al. (2003) surveyed all the dialysis centers in Lebanon to study the effect of consanguineous marriages and their impact on the repartition of kidney diseases and on the risk for familial nephritis. Diabetes, polycystic kidney disease (PKD), chronic pyelonephritis and nephrosclerosis (NS) were the most commonly documented diagnoses in 925 patients reviewed.

Al-Mawi et al (2006) studied the association of HLA DRB1 and DQB1 alleles with type 2 diabetes (T2DM) in Lebanese subjects.

[See also: Bahrain > Nemr et al., 2010].

As part of the WHO ad hoc Diabetes Reporting Group, King and Rewers (1991) studied the prevalence of diabetes mellitus and impaired glucose tolerance among adult communities worldwide. The study revealed that total glucose tolerance was prevalent at a frequency of 30% in the Omani Arab population, as compared to between 11 and 20% in European and U.S. White populations. King and Rewers (1991) were of the opinion that an apparent epidemic of diabetes mellitus was occurring in the adult population across the world, and especially in developing countries.

El-Haddad and Saad (1998) conducted a study to estimate the prevalence of diabetic retinopathy and to determine the medical risk factors. A total of 500 diabetics, including 404 patients with type 2 diabetes who attended a diabetes clinic in Oman between the period of 1996 and 1997 were interviewed and clinically examined. All patients were ophthalmologically examined and were graded according to the findings of the worst eye as grade zero (no retinopathy), grade 1 (mild retinopathy), grade 2 (moderate-severe retinopathy), and grade 3 (proliferative retinopathy). Their glycemic control was classified as good, fair or poor according to their fasting capillary glucose. Out of the 500 diabetic patients, 212 (42.4%) were found to have diabetic retinopathy, with mild non-proliferative retinopathy (NPR) in 25.6%, 4% had moderate to severe NPR, and 12.8% had proliferative retinopathy. The mean age of those with retinopathy was 36.9 years and that of those without retinopathy was 40.2 years. Although the difference was not big, a significant relationship between age and development of retinopathy was found in this study as the risk of retinopathy was more in those aged 40-years or less than in those aged more than 40-years and with every decrease in patient's age by 1 SD, the risk increased by 1.6 times. On the other hand, no relationship between sex and retinopathy was detected. As regarding the type of diabetes, retinopathy was detected in 176 out of 404 (83%) patients with NIDDM when compared to those with IDDM with retinopathy (36 out of 96 patients), and all types of retinopathy were more prevalent in NIDDM than in IDDM (23.2% vs. 2.4% for mild NPR, 3.2% vs. 0.8% for moderate - severe NPR, and 8.8% vs. 4% for proliferative retinopathy). The duration of diabetes for 10 years or more was associated with 8.7 fold increase of the risk of developing retinopathy regardless of the type of treatment of diabetes. Other risk factors identified by this study were the presence of IHD (28 with IHD had retinopathy when compared to only 16 without retinopathy), Systolic Blood Pressure more than 140 mmHg (mean SBP of those with retinopathy was 150 mmHg when compared to 125 mmHg in those without retinopathy), Diastolic Blood Pressure more than 90mmHg, and fasting capillary glucose (mean of 9.8 mmol/l in patients with retinopathy). As regarding the laboratory results, blood urea, creatinine, cholesterol and triglycerides were all significantly related to development of retinopathy, with triglycerides carrying the maximum risk. The multiple logistic regression analysis with adjustment of confounders, revealed that age, IHD, high SBP, high FCG, and high urea were no longer significant, but longer duration of diabetes, high levels of blood cholesterol, triglycerides, high creatinine, and high DBP were still significant risk factors for the development of diabetic retinopathy. With further analysis of these significant risk factors, the duration of diabetes was associated with development of mild NPR, while the other types of retinopathy were associated with all the factors (except for duration of diabetes which was not significant for the occurrence of proliferative retinopathy).

Al-Asfoor et al. (1999) studied the role of body fat distribution as a predictor of NIDDM in the Omani population. Data (with exclusion of pregnant ladies) was obtained from the National diabetic Survey which was a cross sectional study conducted between 1991 and 1992, including more than 4700 subjects above the age of 20 years. The diagnosis of diabetes was made after an oral glucose tolerance test with two hour post load serum glucose cut-off value of 11.1 mmol/l. Waist circumference was measured at the lowest point of the costal margins, and the hip circumference was taken at the widest point of the hip, and the waist/hip ratio (WHR) as well as the waist circumference were divided into six percentiles (10th, 25th, 50th, 75th, 90th, and more than 90th), while the BMI which was the indicator of obesity was divided into four categories (25 kg/m2, 25-29.9 kg/m2, 30-34.9 kg/m2, and 35 kg/m2). The confounding variables were physical activity, family history of diabetes in the immediate family members, and systolic and diastolic blood pressures. Logistic regression was used for univariate and multivariate analysis. Univariate analysis revealed that NIDDM was strongly associated with all measures of obesity ( BMI, waist circumference and WHR), as the lowest risk of NIDDM was seen with low BMI of less than 25 kg/m2 when compared to 35 kg/m2 (odds ratio of 3.5, 95% CI of 2.4 to 4.9). As regarding the WHR, the highest risk was associated with high ratio of 0.97 when compared to a ratio of less than 0.79 (odds ratio=17.9, 95% CI of 7.7 to 41.3). Multivariate analysis revealed a strong association between BMI and NIDDM after adjustment of age, sex, physical activity, family history of diabetes, and blood pressure. On the other hand with adjustment of the above factors in addition to BMI, the association between NIDDM and WHR remained strong with the odds ratio increasing with the increase of the WHR. This would reflect that the prevalence of NIDDM would increase with increasing WHR as those in the highest WHR category had six times the prevalence of NIDDM when compared with those in the lowest category and with comparable BMI levels. The waist circumference was found to be a significant predictor of NIDDM, as those with a circumference of 102 cm or more had a prevalence of diabetes of more than 70%. This was not the case for BMI with adjustment of WHR, which revealed reduced significant association with NIDDM and absent association was detected with adjustment of the waist circumference (making it a non independent risk factor). Al-Asfoor et al. (1999) concluded that WHR and waist circumference were independent factors for developing NIDDM besides BMI, with the waist circumference being the strongest and simplest predictor.

Al-Lawati et al. (2002) conducted a nation-wide survey during year 2000 to determine the prevalence of diabetes in relation to age, gender and region, and compared these results with those obtained from reanalysis of the 1991 survey. They also estimated the prevalence of overweight and obesity as well as the percentage of undiagnosed diabetes. The sample which was collected by multi-stage stratified probability sampling technique had all subjects of Omani nationality and were 20 years or older. They were members of households which were selected from a sampling frame made from a census of all households within 49 randomly selected Census Enumeration Areas which gave urban to rural ratio of 2:1 (similar to the national census). These areas were selected from 16 counties (which were randomly selected from 59 counties of Oman). Out of 7015 eligible subjects, 5838 were interviewed and taken blood samples from (response rate of 83%). The prevalence rate of diabetes was 11.8% and 11.3% in males and females, respectively, while the prevalence of IFG was 7.1% and 5.1% in males and females, respectively. The proportion of subjects with diabetes exceeded 20% in both genders by the age of 50 years. In the age group of 30-64 years, there were 1447 males and 1535 females, in which the crude prevalence of diabetes was 16% and 15.4%, respectively, but with age adjustment the figures increased slightly to 16.3% and 16%, respectively which showed a marked increase from those obtained from 1991 survey (12.8% and 11.9% in males and females, respectively). IFG prevalence became 9.3% and 7.2% in males and females, respectively and age adjustment had not affected it. The overall prevalence of diabetes and IFG in the ten administrative region of Oman ranged from 8% to 18% and 2% to 8%, respectively, with the capital of Muscat having the highest combined prevalence of both diabetes and IFG of 26%, followed by Adh-Dhahirah (22%) and Dhofar (21%) regions. This reflected that diabetes tended to be more common in the urban areas than in the rural ones (urban to rural ratio between prevalence rates was 235:100). About two-thirds of subjects with diabetes were undiagnosed, as only one third of the 572 subjects with positive blood tests self reported diabetes. Overweight (BMI more than 25 but less than 30) was seen in 28.9% of the population and 18.5% were obese (BMI more than 30). Al-Lawati et al. (2002) expected that this prevalence of diabetes would further increase as Oman would show further demographic and socio-economic changes, and that in order to control diabetes mellitus in Oman, primary prevention measures through life style modifications should be undertaken.

Al-Riyami and Afifi (2003) conducted a community based survey to study the accuracy of self reporting of diabetes mellitus (DM) among Omani subjects as compared with the diagnosis of the disease according to the pre-set criteria by using Cohen Kappa, as well as the influence of certain characteristics on this accuracy by using logistic regression. A total of 1968 households representing 16 governorates with 7011 subjects aged >20 years were included, with a response rate of 77.5-91.5% according to physical and laboratory measurement. Household health status questionnaire was used to collect data about DM from 5431 subjects and its prevalence was estimated by summing up those who self reported (4.2%) and those with fasting blood glucose of >7 mmol/l. The prevalence was calculated to be at 11.2%. The kappa was found to be 0.56 for the whole sample, 0.5 for male and 0.61 for females and was highest among the elderly, >60years (0.63) which reflects reporting of inaccurate DM by the male gender and the youngest age group. Logistic regression revealed that those who were >40 years, females, living in urban areas, centrally obese and hypertensive were more likely to report DM, while those with high cholesterol levels were less likely to report it accurately. It was also found that kappa statistics of self reporting of DM was higher than that of hypertension. According to the results obtained from this survey, Al-Riyami and Afifi (2003) concluded that depending on self reporting of DM would give false prevalence rates and that clinical ascertainment in any population-based epidemiological study should be included.

Khandekar et al. (2003) screened subjects with diabetes for their ocular profile and reported the prevalence of diabetic retinopathy as well as its determinants and treatment. This was done by a cross-sectional hospital based descriptive study on 2650 diabetics, including 1806 patients with Type 2 diabetes, who were randomly selected from 5564 subjects screened for diabetic ocular changes. The sample was further subdivided according to the region from which the patients came. The data collected included age, gender, referring institute, duration of diabetes, blood sugar, level of HbAIc, and associated complications of diabetes (hypertension, nephropathy, neuropathy, hyperlipidemia, and coronary artery disease). All patients had their vision determined in each eye, and then their anterior ocular segment, ocular tension and the retina were examined. In case of positive findings of pre-proliferative retinopathy, proliferative retinopathy or maculopathy, laser treatment was proposed. In this study, WHO criteria for diagnosing diabetic retinopathy, vision and visual disability were used. Out of the 2650 selected sample, 2520 (90%) were examined by ophthalmologists, and in 2249 (85%) the ocular media was clear to see the retinal details. There were variations in age, sex, and regions from which the examined sample came, as 60.2% were females while 39.8% were males, most of the patients were in the age group of 40 to 49 years (31%), and most cases were from the capital, Muscat, (21.3%) and North Batinah region (21.3%). The prevalence of diabetic retinopathy was found to be 14.4%. Out of the 1806 patients with NIDDM, 271 (15%) were found to have retinopathy. Retinopathy was found to be significantly higher in males (18.46%) than in females (10.2%), and it was seen more in the age groups of 60 to 69 years (22.87%) and in those above 70 years (15.6%). As regarding the types of retinopathy, 218 patients (8.6%) had background retinopathy, 67 subjects had proliferative retinopathy and 129 subjects (5.1%) had maculopathy with background retinopathy, which was significantly higher than those with maculopathy only without retinopathy (1.3%). There was a direct relationship between the development of retinopathy and the duration of diabetes (interval between first diagnosis of diabetes and this study) as the prevalence of retinopathy was significantly higher in those with diabetes for 11 to 15 years (35.7%) and in those with diabetes for more than 16 years (47.1%) when compared to those with shorter duration of diabetes. When determining the association between glycemic control and retinopathy, 28.3% of those with poor control (HbAIc equals to 9%) had retinopathy when compared to 12.1% of those with better control (HbAIc less than 9%), and it was noted that retinopathy was present in higher frequencies in those with other diabetic complication such as hypertension, neuropathy and nephropathy. As regarding treatment of retinopathy, laser therapy was recommended for 25 subjects (50 eyes) but only five patients (10 eyes) were treated.

Al-Bahrani et al. (2004) determined the sensitivity and specificity of the American diabetes association (ADA) criteria compared with the World Health Organization (WHO) criteria for diagnosis of abnormal glucose intolerance in Omani subjects by measuring their fasting glucose levels and two hour post-75gram oral glucose tolerance test (OGTT) as well as their body mass index (BMI). A total of 176 OGTTs were carried out from 1999 to 2001 on subjects not previously diagnosed with diabetes mellitus. On comparison of the two criteria, it was found that out of the 115 with normal glucose tolerance (NGT 7.8mmol/l), 104 had normal fasting glucose (NFG <6.1mmol/l), while out of the 38 with impaired glucose tolerance (IGT 7.8-11mmol/l), 4 had impaired fasting glucose (IFG 6.1-7mmol/l), and out of the 23 with diabetic glucose tolerance (DGT .11.1mmol/l), 14 had diabetic fasting glucose (DFG .7.0mmol/l). The ADA criteria had missed 90% of subjects with IGT and 39% with DGT. The corresponding sensitivity percentages for ADA and WHO criteria were 90% (NGT), 10% (IGT), and 61% (DGT), while the ADA had 30% sensitivity and 90% specificity when compared to WHO, with likelihood negative and positive ratio of 1.28 and 3, respectively, when all subjects were pooled. The receiver-operating characteristic (ROC) curve had shown that 5.9 mmol/l was the diagnostic cut-off value of fasting glucose for IGT (according to WHO criteria) with the highest sensitivity and specificity when compared to 6.1mmol/l of the ADA criteria. Although the average BMI of the IGT group (mean 31.2, SD 1.2) was found to be higher than the other two groups, it was not statistically significant. Al-Bahrani and colleagues (2004) recommended using the OGTT as a screening test for the high risk group in Oman, as to the high prevalence of diabetes, rather than the fasting glucose, though in the clinical setting , it might have a role in diagnosis of diabetes. In 2005, Al-Bahrani et al. assessed the dyslipidemia in Omani subjects with diabetes type 2 and studied the possibility of using apoB levels in determining the risk of developing cardiovascular disease. The study was conducted on 221 diabetics (diagnosed according to WHO criteria) between the ages of 30 and 70 years who were followed up between 1999 and 2002. History regarding smoking, alcohol, exercise, and prophylactic drugs (beta-blockers, aspirin, diuretics, sulphonylurea) was taken. Management was by diet alone in 40% and the rest were on oral hypoglycemic agents. The control group was made up of 67 healthy age-matched randomly selected subjects. All subjects were clinically examined. High risk categories for IHD were identified if their apoB level was more than 1.2 g/L and their apoA-1 was less than 1.2 g/L, and selection of these levels was because they identified subjects with LDLc concentration of more than 4.2 mmol/l and HDLc concentration of less than 0.92 mmol/l. Al-Bahrani et al. (2005) found that diabetics had significantly higher levels of triglycerides, Tc/HDLc, non-HDLc, apoB/LDLc ratio, and non-significant higher levels of apoB, but lower level of HDLc and non significant lower levels of Lp(a) than in non-diabetics. They also found that the probability in diabetics to have high Tg and low HDLc levels was three to four and six to seven times higher, respectively, than non-diabetics. As regarding the assessment of lipid parameters in diabetics, the Tg correlated positively with apoB and apoB/LDLc ratio, and negatively with apoB/non-HDLc. On the other hand, Lp(a) correlated negatively with Tg and apoB/LDLc in the diabetics, while the correlation was positive with apoB/LDL and apoB/non-HDL cholesterol in the non diabetics. In comparison to non diabetics, diabetics had a lower probability of having LDLc > 4.2 mmol/l and higher probability of having apoB level > 1.2 g/L. At the cut-off value of apoB of more than 2.1 g/l, both groups had similar proportion of subjects with levels of LDLc > 4.2 mmol/l (specificity of 93%), but on the other hand, 60% of diabetics with apoB level > 2.1 g/l had low level of LDLc < 4.2 mmol/l, when compared to 7% of non diabetics (sensitivity of 40% versus 93%). In the case of apoA-1, a value > 1.2 g/l was similar, as both groups had low levels of apoA-1 levels despite having high levels of HDLc (>1.0 mmol/l). As regarding diabetics with Ischemic Heart Disease (IHD), significantly higher levels of apoB/non-HDLc were found on comparison with those with no IHD, and this relationship of IHD with apoB/non-HDLc remained strong even after adjustment of age, sex, BMI, hypertension, lipid parameters, and smoking status.

Al-Moosa et al. (2006) conducted a study to determine the prevalence of diabetes and the associated coronary risk factors and tested whether urbanization along with other risk factors played a role in the development of diabetes. The study sample was taken from the 2000 National Health Survey and included 2044 randomly selected occupied households from randomly selected areas. The capital region of Muscat was considered the urban area while the rest of regions were the rural ones. Subjects who were 20 years or older (7179) were interviewed from 1968 households, and their clinical profile studied. The prevalence of diabetes in the population aged 20 years or older was 11.6% which varied according to the urban or rural residence (17.7% and 10.5%, respectively), as well as other demographic factors, measures of obesity, cholesterol and SBP. Less than half of the subjects who were diagnosed with diabetes by blood tests, self reported to have diabetes and 11.1% of those who did not know that they had diabetes were from the urban area when compared to only 6.2% from the rural areas. The determined prevalence of hypertension and obesity were 21.5% and 19.1%, respectively, but that of high cholesterol was 50.6%. Those residing in the urban area constituted the greater proportion of those with diabetes (17.7% vs. 10.5% in rural), and hypertension (26.4% vs. 20.2%), but obesity and high cholesterol were similarly distributed between urban and rural areas. Initial analysis revealed that illiterate and less educated groups were more likely to develop diabetes than the highly educated subjects. Further analysis of the 5847 out of 7179 subjects (due to lack of information) for the association between diabetes and significant variables revealed that urban dwellers (were twice likely to develop diabetes than rural ones), advanced age (every five year increase in age was associated with 1.2 greater odds of developing diabetes), being married, and abnormal waist circumference increased the odds of diabetes. A strong gradient but not a linear relationship between SBP and diabetes was detected and with increasing SBP from 120 mmHg to 140 mmHg or more, the prevalence of diabetes increased for 6.2% to 19.6% and 24.4% in those with SBP of more than 160 mmHg. Education lost its significance with adjustment of other factors but regained it in the stratified analysis by region as higher educated subjects in the rural areas were less likely to develop diabetes than the illiterate or less educated subjects.

Al-Lawati and Barakat (2007) determined optimal fasting plasma glucose (FPG) cut-point that predicted diabetes diagnosed by OGTT among Omani subjects and validated this value by testing it on an independent population. The data for determination of the cut-point was obtained from the 1991 National Diabetes Survey of Oman (4917) in which the response rate was 80%, and the validation data was obtained from the 2001 Nizwa Survey (1439) in which the response rate was 75%. In both data analyses, candidates were asked to fast for eight to 14 hours after which an OGTT was done with 75 grams of glucose (with exclusion of those who were already diabetics) to identify the diabetics according to the WHO criteria. They were also interviewed regarding behavioral risk factors, medical history, and were clinically examined. The optimal FPG was estimated from plotting a series of receiver-operating characteristic (ROC) curves with the true positive rate (sensitivity) on the y-axis and the false positive rate (1-specifity) on the x-axis, and the cut off point was the one with the highest sensitivity and lowest false positive rate. The larger the area under the curve, the more predictive the test value of the reference would be. Positive and negative likelihood ratios were calculated for each cut-point. Upon comparing the 1998 WHO and the 1997 ADA criteria for diagnosing diabetes, Al-Lawati and Barakat (2007) found that ADA criteria had underestimated diabetes in this Omani population by 18%, as the WHO criteria had diagnosed 489 (9.9%) diabetics, while ADA had diagnosed 397 (8.1%) diabetics, and around 32% of the diabetics according to the WHO criteria had their FPG in the non-diabetic range (less than 7 mmol/l) according to the ADA criteria. The sensitivity and specificity of ADA criteria was found to be 68.3% and 98.6%, respectively. The optimal FPG for diagnosing diabetes determined from the ROC curves were > 6.0mmo/l for males and for non-obese subjects with BMI less than 30 kg/m2, 5.9 mmol/l for females as well as for the overall population and for obese subjects (BMI more than 30 kg/m2). The areas under the two genders ROC curves as well as that under the BMI related analysis were not statistically different. For the validation analysis, the optimal FPG value of more than 5.9 mmol/l was used. In the Nizwa Survey, out of the 1439 subjects, 139 (9.7%) were diagnosed as diabetics according to the WHO criteria and 117 (8.1%) were diagnosed according to the predicted cut off point of > 5.9 mmol/l. The sensitivity and specificity of this cut-point was 84.2% and 80.2%, respectively, when compared to the 60.4% and 96.6%, respectively of the ADA criteria used in Nizwa. An interval of > 5.2 mmol/l and < 5.8mmol/l as FPG concentrations was tested for diagnosing subjects with impaired glucose tolerance, but as its sensitivity and specificity were low (27.1% and 74.5%) it could not be used in practice.

El-Tonsy et al. (1998) studied 87 patients (33 Qatari and 54 other nationalities; 73 males and 14 females;76 adults and 11 children; aged 2-56 years) with alopecia areata (AA). They found six patients (6.9%) to have diabetes mellitus, and 27 patients (31.1%) were found to have a family history for diabetes mellitus. [El-Tonsy MH, Azadeh B, Kamal AM, El-Domyati MBM, Ibrahim FA. Auto antibodies and immunohistochemical studies in alopecia areata. Gulf J Dermatol. 1998; 5(1): 40-45.]

Bener et al. (2005) studied the association between consanguineous marriages, and genetic and environmental factors, and NIDDM in the adult Qatari population. A total of 338 diabetic patients (mean age: 45.4 years) were selected at random and compared to 344 healthy control subjects (mean age: 42.4 years). Family histories and social conditions of the patients and control subjects were studied, and height, weight, blood pressure, serum insulin and glucose levels were measured. The diabetic population was found to have a significantly lower educational level and a tendency to live in houses with less than five rooms. Diabetics were significantly higher among the subjects with consanguinity, and this effect was even more significant in first degree consanguinity. Diabetics were also seen to be more obese and to consume more fruit and less fish/chicken. Obesity, mean BMI, systolic blood pressure, total cholesterol, HDL-cholesterol, and serum triglycerides were also higher among diabetics. Logistic regression analysis identified smoking, consanguinity, BMI, level of education, number of children, and systolic blood pressure as risk factors to be considered for NIDDM. Bener et al. (2005) suggested that the high prevalence of diabetes in the Qatari population could be attributed to an interaction of environmental and genetic factors.

Daghash et al. (2007) conducted a prospective study during the period from May to October 2004 at Hamad General Hospital, aiming to assess the lipoprotein profile in Qatari patients with type 2 diabetes mellitus (T2DM) and to assess its relationship to coronary artery disease (CAD). The study subjects included Qatari nationals, 180 cases with diabetes (43.3% males, and 56.7% females) and 180 controls without diabetes (52.2% males, and 47.8% females), aged 25 to 65 years. Diabetic patients were further divided to 57 cases with CAD and to 123 cases without CAD. Consanguinity, obesity, total cholesterol, reduced HDL-cholesterol and triglyceride were more prevalent in diabetic patients. Also, there was statistically significant difference in biochemistry levels between diabetic patients with CAD and without CAD for age, gender, serum creatinine, triglycerides and BMI. The patients' LDL correlated significantly with total Cholesterol and Lp B. Total cholesterol correlated significantly with triglycerides, Lp B and Lp (A). Daghash et al. (2007) concluded that Lp(a) may not be an independent risk factor for CAD in patients with DM.

Badii et al. (2008) carried out a matched case-control study among diabetic patients and healthy subjects. This survey was conducted from February 2003 to March 2006 in Qatari male and female nationals aged 35 to 60 years. The study was based on matched age, sex, and ethnicity of 400 cases (with diabetes) and 450 controls (without diabetes). Face-to-face interviews were based on a questionnaire that included variables such as age, sex, sociodemographic status, body mass index (BMI), and obesity. Their health status was assessed by medical conditions, family history, and blood pressure measurements. Nearly half of the diabetic type 2 patients (48.5%) were found obese (BMI > 30) compared with non-diabetic subjects (29.8%) (P < 0.001). The prevalence of T2D was found significantly higher among the consanguineous group (P < 0.001). Furthermore, it was observed that the chances of diabetes were even higher among couples of first- and second-degree of consanguinity. Systolic and diastolic blood pressure levels were found significantly higher in diabetes patients than in non-diabetic healthy subjects (P < 0.001). Badii et al. (2008) did not find any association between the Pro12Ala polymorphism of the PPARG2 gene and type 2 diabetes in Qatar.

In order to determine the prevalence of diagnosed and undiagnosed diabetes, pre-diabetes, and to identify the associated risk factors in the sample of adult Qatari population, Bener et al. (2009) carried out a cross-sectional study conducted from January 2007 to July 2008 in urban and semi-urban primary health care centers among Qatari nationals above 20 years of age. Of the 1434 subjects who were approached to participate in the study, 1117 (77.9%) gave their consent. The overall prevalence of diabetes mellitus among adult Qatari population was found to be high (16.7%) with diagnosed DM (10.7%) and newly diagnosed DM (5.9%). The impaired glucose tolerance (IGT) was diagnosed in 12.5%, while impaired fasting glucose was in 1.3% with a total of (13.8%). The proportion of DM was found to be higher in Qatari women (53.2%) than in Qatari men (46.8%) and it peaked in the age group 40-49 years (31.2%). The age-specific prevalence of total DM and IGT was found to be increased with age. Risk factors were found to be significantly higher in diabetic adult Qatari population: central obesity (p<0.001), hypertension (p<0.001), triglyceride (p<0.001), HDL (p=0.003), metabolic syndrome (p<0.001), heart diseases (p<0.001). Smoking habits and family history of DM were found to be the major contributors for diabetes disease. The central obesity was found to be associated with higher prevalence of DM and IFG among Qatari men and women. Bener et al. (2009) concluded that high proportion of pre-diabetes in Qatari adults will increase the prevalence of DM in the next few years; smoking habits and family history of DM were the major contributors for DM; and early diagnosis of DM is of major importance to reduce the risk of these diabetes-related conditions.

A study by Alsmadi et al. (2008) was set to determine the role of two variants of the TCF7L2 gene in Arabs in the context of type 2 diabetes (T2D); these variants are rs7903146 and rs12255372.  The latter variants have been strongly associated with T2D risk in many populations around the world.  Saudi T2D patients (522 individuals), and controls (346 individuals) aged over 60, with fasting plasma glucose < 7 mmol/L were included in a case-control association study.  The results indicated weak or no association of T2D in Arabs with the abovementioned TCF7L2 variants.

Alsmadi et al., (2008) carried out a case-control association study using 550 type 2 diabetes Saudi patients aged above 60 years, and 335 age-matched controls.  Genotyping was done using molecular beacon-based real time PCR assays E23K.  The frequency of the K allele among the cases (231; 21%) was significantly higher than that observed among the controls (91; 13.6%).  Also the GA genotype frequency in the patients was significantly higher than that observed in the controls when compared to the GG reference genotype. There was no significant difference between the cases and controls in AA genotype frequency compared to the GG reference genotype.

Al-Daghri et al. (2016) attempted to study the association between SNAP25 polymorphisms and T2DM in the Saudi population.  A total of 489 T2DM affected patients and 530 healthy controls were recruited for the study.  Their anthropometric measurements, fasting blood glucose, serum insulin and lipid profile were noted.  Genomic DNA was isolated and the SNAP25 SNPs rs363043, rs363039 and rs363050 were genotyped.  Statistical analysis studies did not find a significant association between any of the three SNPs and T2DM.  However, T2DM patients carrying the rs363050 AG/GG genotype were found to have significantly higher levels of fasting glucose and HbA1c than patients carrying the AA genotype (p=0.03, p=0.03).  The AG/GG patients also had significantly lower levels of serum insulin compared to AA patients (p=0.009).  The SNAP25 rs363050 G allele was thus not found to be a genetic risk factor for T2DM, but to correlate with a more severe phenotype once T2DM had been diagnosed.  As the rs363050 G allele resulted in a reduced expression of SNAP25, the authors suggested that it may affect the exocytotic machinery and result in the suboptimal release of insulin.  

Al-Daghri et al., (2016) randomly selected 814 Saudi adults, 394 of whom had type 2 diabetes and the rest were controls, from a biomarker screening project in Riyadh.  The aim was to examine the possible role of the following five FDNC5 polymorphisms; rs3480A/G, rs1746661G/T, rs1298190A/G, rs726344A/G and rs1570569G/T on type 2 diabetes and obesity since both were common in Saudi Arabia.  DNA samples from all subjects were genotyped for the 5 SNPs using TaqMan genotyping assay. The authors found that serum Irisin levels were higher in patients with type 2 diabetes compared to controls.  The analyses of the five SNPs showed that rs3480 GG is associated with decreased risk of obesity and low body mass index.  The rs1746661 G allele is associated with high levels of triglyceride.  The rs157069 TT genotype is associated with high levels of fasting insulin and HOMA-IR while it was associated with lower levels of circulating Irisin.  The authors suggested that further analyses were needed to understand the underlying mechanisms.

Ahmed et al. (1999) studied 854 diabetic patients attending a weekly clinic in Medani in a 26 month period, and found that seven of these patients had intermittent watery diarrhea. These patients, however, had no evidence of malabsorption or any nutritional deficit.

[Ahmed AM, Ahmed AM, Ahmed NH. Diabetic diarrhea in Sudanese patients. Qatar Med J. 1999; 8(2):45-6.]

Zouari Bouassida et al. (2004) described associations between type 2 diabetes polymorphisms in the promoter region of the Tumor Necrosis Factor alfa (TNF-alpha) and in the gene of Heat-shock protein 70-2 (HSP70-2). With polymerase chain reaction and restriction enzyme, they examined 280 type 2 diabetes unrelated Tunisian patients and 274 control subjects. The heterozygous TNF1/TNF2 genotype of the TNF-alpha gene is suggested as a protective marker against type diabetes whereas the homozygous P2/P2 of the HSP70-2 gene may represent an increased risk for type diabetes independent from the obesity present in some of the patients.

Omer et al. (1985) conducted a retrospective analysis on patients with diabetes observed in Al-Ain, the second largest city in the Abu Dhabi Emirate, between 1980 and 1984. Omer et al. (1985) indicated that diabetes accounted for 6% of all general medical admissions to hospitals in the city. [Omer A, Elsir K, Muneer M. Diabetes mellitus in Al-Ain: the impact on hospital services. Emirate Med J 1985; 3:119-22.]

In 1995, El Mugamer et al. conducted a study coronary heart disease risk factors of non-insulin dependent diabetes mellitus (NIDDM), obesity and hypertension using community based survey among a bedouin-derived Emirati population sample of 322 subjects (> or = 20 years). Overall diabetes prevalence was 6% (11% in male and 7% in female subjects aged 30-64 years).