Factors associated with hypoglycemia episodes in hospitalized type 2 diabetes mellitus patients in a tertiary health facility in Malaysia

Purpose: To determine the factors associated with severity of hypoglycemia in hospitalized type 2 diabetes mellitus patients in a tertiary health facility in Malaysia. Methods: This retrospective study involved 207 hospitalised T2DM patients with hypoglycaemia episodes from January 2008 to December 2012 and was conducted in University Malaya Medical Centre, Petaling Jaya, Malaysia. Patients were classified into 2 groups, viz, those who had hypoglycaemia on admission and those who had hypoglycaemia during hospital stay. Patients with hypoglycemia on admission were those admitted due to hypoglycemia while patients with hypoglycemia during hospital stay were those admitted due to other causes but subsequently developed hypoglycemia during hospitalization. Results: The results for the 207 patients investigated show that most of the patients (72.2 %) were asymptomatic during hypoglycemic episodes. The majority of the episodes (57.4 %) experienced by the patients were mild hypoglycemia (< 3.9 mmol/L). Old age (p = 0.011) and presence of stroke (p = 0.033) were found to be significantly associated with severe hypoglycemia (< 2.2 mmol/L) while concurrent use of opioid (p = 0.008) was associated with mild hypoglycemia. Conclusion: The identification of the underlying factors associated with severity of hypoglycemia may help in preventing and resolving hypoglycemia in T2DM patients.

Training of interval type-2 fuzzy logic system using extreme learning machine for load forecasting

Extreme learning machine (ELM) is originally proposed for single- hidden layer feed-forward neural networks (SLFN). From the functional equivalence of fuzzy logic systems and SLFN, the fuzzy logic systems can be interpreted as a special case of SLFN under some mild conditions. Hence the fuzzy logic systems can be trained using SLFN's learning algorithms. Considering the same equivalence, ELM is utilized here to train interval type-2 fuzzy logic systems (IT2FLSs). Based on the working principle of the ELM, the parameters of the antecedent of IT2FLSs are randomly generated while the consequent part of IT2FLSs is optimized using Moore-Penrose generalized inverse of ELM. Application of the developed model to electricity load forecasting is another novelty of the research work. Experimental results shows better forecasting performance of the proposed model over the two frequently used forecasting models.

Mortality trends among people with type 1 and type 2 diabetes in Australia: 1997-2010

OBJECTIVE: With improvements in cardiovascular disease (CVD) rates among people with diabetes, mortality rates may also be changing. However, these trends may be influenced by coding practices of CVD-related deaths on death certificates. We analyzed trends of mortality over 13 years in people with diabetes and quantified the potential misclassification of CVD mortality according to current coding methods. RESEARCH DESIGN AND METHODS: A total of 1,136,617 Australians with diabetes registered on the National Diabetes Services Scheme between 1997 and 2010 were linked to the National Death Index. Excess mortality relative to the Australian population was reported as standardized mortality ratios (SMRs). Potential misclassification of CVD mortality was determined by coding CVD according to underlying cause of death (COD) and then after consideration of both the underlying and other causes listed in part I of the death certificate. RESULTS: For type 1 diabetes, the SMR decreased in males from 4.20 in 1997 to 3.08 in 2010 (Ptrend < 0.001) and from 3.92 to 3.46 in females (Ptrend < 0.01). For type 2 diabetes, the SMR decreased in males from 1.40 to 1.21 (Ptrend < 0.001) and from 1.56 to 1.22 in females (Ptrend < 0.001). CVD deaths decreased from 35.6 to 31.2% and from 31.5 to 27.2% in males and females with type 1 diabetes, respectively (Ptrend < 0.001 for both sexes). For type 2 diabetes, CVD decreased from 44.5 to 29.2% in males and from 45.5 to 31.6% in females (Ptrend < 0.001 for both sexes). Using traditional coding methods, ∼38 and 26% of CVD deaths are underestimated in type 1 diabetes and type 2 diabetes, respectively. CONCLUSIONS: All-cause and CVD mortality has decreased in diabetes. However, the total CVD mortality burden is underestimated when only underlying COD is considered. This has important ramifications for understanding mortality patterns in diabetes.

Explaining psychological insulin resistance in adults with non-insulin-treated type 2 diabetes: the roles of diabetes distress and current medication concerns. Results from diabetes MILES-Australia.

AIMS: To investigate the contribution of general and diabetes-specific emotional wellbeing and beliefs about medicines in the prediction of insulin therapy appraisals in adults with non-insulin-treated type 2 diabetes. METHODS: The sample included Diabetes MILES-Australia cross-sectional survey participants whose primary diabetes treatment was oral hypoglycaemic agents (N=313; 49% women; mean±SD age: 57±9 years; diabetes duration: 7±6 years). They completed validated measures of beliefs about the 'harm' and 'overuse' of medications in general (BMQ General); 'concerns' about and 'necessity' of current diabetes medications (BMQ Specific); negative insulin therapy appraisals (ITAS); depression (PHQ-9); anxiety (GAD-7), and diabetes distress (DDS-17). Factors associated with ITAS Negative scores were examined using hierarchical multiple regressions. RESULTS: Twenty-two percent of the variance in ITAS Negative scores (52±10), was explained by: number of complications (β=-.15, p=.005), DDS-17 subscale 'emotional burden' (β=.23, p<.001), and 'concerns' about current diabetes treatment (β=.29, p<.001). General beliefs about medications and general emotional wellbeing did not contribute significantly to the model. CONCLUSIONS: Psychological insulin resistance may reflect broader distress about diabetes and concerns about its treatment but not general beliefs about medicines, depression or anxiety. Reducing diabetes distress and current treatment concerns may improve attitudes towards insulin as a potential therapeutic option.

The contribution of workplace characteristics to the risk of Type 2 Diabetes

Diabetes has a significant economic impact on individuals, families, health systems, and countries [1]. In 2010 it was estimated that the global health expenditure on diabetes was US376 billion (€292 billion), equating to 12% of health expenditure and US1330 (€1031) per person [2]. A separate estimate in 2010 reported that diabetes cost the US US174 billion (€135 billion), with US58 billion (€45 billion) in indirect costs equating to over US2000 (€1551) on average per person with diabetes [3]. The World Health Organization estimates that between 2005 and 2030 the proportion of deaths caused by diabetes will double and global health expenditures associated with diabetes are expected to reach US490 billion (€380 billion) [1,2]. In 2003, it was estimated that diabetes cost Australia AUS6 billion (US6 billion, €5 billion), with AUS21 million (US22 million, €17 million) in indirect costs such as lost workdays and lost productivity equating to approximately AUS35 (US35, €28) per person with diabetes [4].

Retinal tissue thickness in type 1 and type 2 diabetes

BACKGROUND: The objective was to investigate full retinal and inner retinal thickness in individuals with type 1 and type 2 diabetes. METHODS: Eighty-four individuals with type 1 diabetes (T1DM), 67 individuals with type 2 diabetes (T2DM) and 42 non-diabetic individuals (control group) were enrolled. Participants underwent full retinal thickness evaluation in the central retinal, parafoveal and perifoveal zones and in the retinal nerve fibre layer (RNFL) and ganglion cell complex (GCC), using spectral domain optical coherence tomography. As a preliminary step, the key variables of interest - age, sex, diabetic retinopathy (DR), duration of diabetes and HbA1c levels - were analysed and compared between the three groups. Full retinal thickness, RNFL and GCC thicknesses were also compared between the groups. The relationship between the type of diabetes and retinal tissue thickness was explored, adjusting for the five potential confounders. RESULTS: Compared to individuals with T1DM, individuals with T2DM had significantly reduced full retinal thickness in the parafovea and perifovea and reduced RNFL and GCC thickness. The mean differences were six (p = 0.020), seven (p = 0.008), six (p = 0.021) and four micrometres (p = 0.013) for the parafovea, perifovea, RNFL and GCC thicknesses, respectively. Thicknesses within the central zone (p = 0.018) and at the parafovea (p = 0.007) were significantly reduced in T2DM when compared to the control group. After adjusting for age, sex, diabetic retinopathy, duration of diabetes and HbA1c levels, the relationship between type of diabetes and retinal tissue thickness was not statistically significant (p > 0.056). CONCLUSION: Retinal tissue thickness is not significantly different between type 1 and type 2 diabetes, when adjusted for age, sex, diabetic retinopathy, duration of diabetes and HbA1c levels.

Epigenetic markers to predict conversion from gestational diabetes to type 2 diabetes

CONTEXT: Lifestyle factors mediate epigenetic changes that can cause chronic diseases. Although animal and laboratory studies link epigenetic changes to diabetes, epigenetic information in women with gestational diabetes (GDM) and type 2 diabetes is lacking. OBJECTIVE: To measure epigenetic markers across pregnancy and early postpartum and identify markers that could be used as predictors for conversion from GDM to type 2 diabetes. DESIGN: Global histone H3 dimethylation was measured at three time points: 30 weeks gestation, 8-10 weeks postpartum and 20 weeks postpartum, in white blood cells from four groups of women with and without diabetes. SETTING AND PARTICIPANTS: A total of 39 participants (six to nine in each group) were recruited including: non-diabetic women; women with GDM who developed postpartum type 2 diabetes; women with GDM without postpartum type 2 diabetes; and women with type 2 diabetes. MAIN OUTCOME MEASURE: Percentages of dimethylation of H3 histones relative to total H3 histone methylation were compared between diabetic/non-diabetic groups using appropriate comparative statistics. RESULTS: H3K27 dimethylation was 50-60% lower at 8-10 and 20 weeks postpartum in women with GDM who developed type 2 diabetes, compared with non-diabetic women. H3K4 dimethylation was 75% lower at 8-10 weeks postpartum in women with GDM who subsequently developed type 2 diabetes compared with women who had GDM who did not. CONCLUSIONS: The percentage of dimethylation of histones H3K27 and H3K4 varied with diabetic state and has the potential as a predictive tool to identify women who will convert from GDM to type 2 diabetes.

Hybrid model for the training of interval type-2 fuzzy logic system

In this paper, a hybrid training model for interval type-2 fuzzy logic system is proposed. The hybrid training model uses extreme learning machine to tune the consequent part parameters and genetic algorithm to optimize the antecedent part parameters. The proposed hybrid learning model of interval type-2 fuzzy logic system is tested on the prediction of Mackey-Glass time series data sets with different levels of noise. The results are compared with the existing models in literature; extreme learning machine and Kalman filter based learning of consequent part parameters with randomly generated antecedent part parameters. It is observed that the interval type-2 fuzzy logic system provides improved performance with the proposed hybrid learning model.

Age-specific trends from 2000-2011 in all-cause and cause-specific mortality in type 1 and type 2 diabetes: a cohort study of more than one million people

OBJECTIVE: To analyze changes by age-group in all-cause and cause-specific mortality rates from 2000-2011 in people with diabetes. RESEARCH DESIGN AND METHODS: A total of 1,189,079 (7.3% with type 1 diabetes) Australians with diabetes registered on the National Diabetes Service Scheme between 2000 and 2011 were linked to the National Death Index. Mortality rates in the total population were age standardized to the 2001 Australian population. Mortality rates were calculated for the following age-groups: 0 to <40 years, ≥ 40 to <60 years, and ≥60 to ≤85 years. Annual mortality rates were fitted using a Poisson regression model including calendar year as a covariate and age and sex where appropriate, with Ptrend reported. RESULTS: For type 1 diabetes, all-cause, cardiovascular disease (CVD), and diabetes age-standardized mortality rates (ASMRs) decreased each year by 0.61, 0.35, and 0.14 per 1,000 person-years (PY), respectively, between 2000 and 2011, Ptrend < 0.05, while cancer mortality remained unchanged. By age, significant decreases in all-cause, CVD, and diabetes mortality rates were observed in all age-groups, excluding diabetes mortality in age-group 0-40 years. For type 2 diabetes, all-cause, CVD, and diabetes ASMRs decreased per year by 0.18, 0.15, and 0.03 per 1,000 PY, respectively, Ptrend < 0.001, while cancer remained unchanged. By age, these decreases were observed in all age-groups, excluding 0-40 years, where significant increases in all-cause and cancer mortality were noted and no change was seen for CVD and diabetes mortality. CONCLUSIONS: All-cause, CVD, and diabetes ASMRs in type 1 and type 2 diabetes decreased between 2000 and 2011, while cancer ASMRs remained unchanged. However, younger populations are not benefiting from the same improvements as older populations. In addition, the absence of a decline in cancer mortality warrants urgent attention.

Can the direct medical cost of chronic disease be transferred across different countries? Using cost-of-illness studies on type 2 diabetes, epilepsy and schizophrenia as examples

OBJECTIVES: To systematically review cost of illness studies for schizophrenia (SC), epilepsy (EP) and type 2 diabetes mellitus (T2DM) and explore the transferability of direct medical cost across countries.

METHODS: A comprehensive literature search was performed to yield studies that estimated direct medical costs. A generalized linear model (GLM) with gamma distribution and log link was utilized to explore the variation in costs that accounted by the included factors. Both parametric (Random-effects model) and non-parametric (Boot-strapping) meta-analyses were performed to pool the converted raw cost data (expressed as percentage of GDP/capita of the country where the study was conducted).

RESULTS: In total, 93 articles were included (40 studies were for T2DM, 34 studies for EP and 19 studies for SC). Significant variances were detected inter- and intra-disease classes for the direct medical costs. Multivariate analysis identified that GDP/capita (p<0.05) was a significant factor contributing to the large variance in the cost results. Bootstrapping meta-analysis generated more conservative estimations with slightly wider 95% confidence intervals (CI) than the parametric meta-analysis, yielding a mean (95%CI) of 16.43% (11.32, 21.54) for T2DM, 36.17% (22.34, 50.00) for SC and 10.49% (7.86, 13.41) for EP.

CONCLUSIONS: Converting the raw cost data into percentage of GDP/capita of individual country was demonstrated to be a feasible approach to transfer the direct medical cost across countries. The approach from our study to obtain an estimated direct cost value along with the size of specific disease population from each jurisdiction could be used for a quick check on the economic burden of particular disease for countries without such data.

A syntenic cross species aneuploidy genetic screen links RCAN1 expression to β-cell mitochondrial dysfunction in Type 2 diabetes

Type 2 diabetes (T2D) is a complex metabolic disease associated with obesity, insulin resistance and hypoinsulinemia due to pancreatic β-cell dysfunction. Reduced mitochondrial function is thought to be central to β-cell dysfunction. Mitochondrial dysfunction and reduced insulin secretion are also observed in β-cells of humans with the most common human genetic disorder, Down syndrome (DS, Trisomy 21). To identify regions of chromosome 21 that may be associated with perturbed glucose homeostasis we profiled the glycaemic status of different DS mouse models. The Ts65Dn and Dp16 DS mouse lines were hyperglycemic, while Tc1 and Ts1Rhr mice were not, providing us with a region of chromosome 21 containing genes that cause hyperglycemia. We then examined whether any of these genes were upregulated in a set of ~5,000 gene expression changes we had identified in a large gene expression analysis of human T2D β-cells. This approach produced a single gene, RCAN1, as a candidate gene linking hyperglycemia and functional changes in T2D β-cells. Further investigations demonstrated that RCAN1 methylation is reduced in human T2D islets at multiple sites, correlating with increased expression. RCAN1 protein expression was also increased in db/db mouse islets and in human and mouse islets exposed to high glucose. Mice overexpressing RCAN1 had reduced in vivo glucose-stimulated insulin secretion and their β-cells displayed mitochondrial dysfunction including hyperpolarised membrane potential, reduced oxidative phosphorylation and low ATP production. This lack of β-cell ATP had functional consequences by negatively affecting both glucose-stimulated membrane depolarisation and ATP-dependent insulin granule exocytosis. Thus, from amongst the myriad of gene expression changes occurring in T2D β-cells where we had little knowledge of which changes cause β-cell dysfunction, we applied a trisomy 21 screening approach which linked RCAN1 to β-cell mitochondrial dysfunction in T2D.

'Are you sure you're going to have another one of those?': a qualitative analysis of the social control and social support models in type 2 diabetes

While there is evidence that spouses can impact the self-management of adults with type 2 diabetes mellitus, less is known about the influence of the wider social network. This qualitative study explored the perceived impact of the family as well as friends and work colleagues on type 2 diabetes mellitus self-management. A total of 25 adults with type 2 diabetes mellitus participated in semi-structured interviews regarding their social experiences of living with diabetes. Deductive thematic analysis was applied to the data. Pre-existing themes of health-related social control and social support were identified in the wider social network, with additional themes of non-involvement and unintentional undermining also emerging.

Modified AHP for gene selection and cancer classification using type-2 fuzzy logic

This paper proposes a modification to the analytic hierarchy process (AHP) to select the most informative genes that serve as inputs to an interval type-2 fuzzy logic system (IT2FLS) for cancer classification. Unlike the conventional AHP, the modified AHP allows us to process quantitative factors that are ranking outcomes of individual gene selection methods including t-test, entropy, receiver operating characteristic curve, Wilcoxon test, and signal-to-noise ratio. The IT2FLS is introduced for the classification task due to its great ability for handling nonlinear, noisy, and outlier data, which are common problems in cancer microarray gene expression profiles. An unsupervised learning strategy using the fuzzy c-means clustering is employed to initialize parameters of the IT2FLS. Other classifiers such as multilayer perceptron network, support vector machine, and fuzzy ARTMAP are also implemented for comparisons. Experiments are carried out on three well-known microarray datasets: diffuse large B-cell lymphoma, leukemia cancer, and prostate. Rather than the traditional cross validation, leave-one-out cross-validation strategy is applied for the experiments. Results demonstrate the performance dominance of the IT2FLS against the competing classifiers. More noticeably, the modified AHP improves the classification performance not only of the IT2FLS but of all other classifiers as well. Accordingly, the proposed combination between the modified AHP and IT2FLS is a powerful tool for cancer classification and can be implemented as a real clinical decision support system that is useful for medical practitioners.