A Federated Database for Obesity Research: An IMI-SOPHIA Study.

Obesity is considered by many as a lifestyle choice rather than a chronic progressive disease. The Innovative Medicines Initiative (IMI) SOPHIA (Stratification of Obesity Phenotypes to Optimize Future Obesity Therapy) project is part of a momentum shift aiming to provide better tools for the stratification of people with obesity according to disease risk and treatment response. One of the challenges to achieving these goals is that many clinical cohorts are siloed, limiting the potential of combined data for biomarker discovery. In SOPHIA, we have addressed this challenge by setting up a federated database building on open-source DataSHIELD technology. The database currently federates 16 cohorts that are accessible via a central gateway. The database is multi-modal, including research studies, clinical trials, and routine health data, and is accessed using the R statistical programming environment where statistical and machine learning analyses can be performed at a distance without any disclosure of patient-level data. We demonstrate the use of the database by providing a proof-of-concept analysis, performing a federated linear model of BMI and systolic blood pressure, pooling all data from 16 studies virtually without any analyst seeing individual patient-level data. This analysis provided similar point estimates compared to a meta-analysis of the 16 individual studies. Our approach provides a benchmark for reproducible, safe federated analyses across multiple study types provided by multiple stakeholders.

Differences in insulin dosing in women with type 1 diabetes before and after the menopause.

The menstrual cycle increases insulin requirements in a subset of women with type 1 diabetes as a result of reduced insulin action through sexual hormones. If exposure to sexual hormones declines during the menopause, adaptions of insulin dosing may be required. However, there are no validated recommendations available on how to adapt insulin treatment in postmenopausal women with type 1 diabetes. The present study compared insulin dosing profiles of 630 premenopausal and 548 postmenopausal women, who had long-term type 1 diabetes and used continuous subcutaneous insulin infusion. Data were extracted from the German "Diabetes-Patienten- Verlaufsdokumentation". It was found that total daily insulin (p <0.0001), daily insulin per kilogram bodyweight (p <0.0001), total daily basal insulin (p <0.0001), daily basal insulin per kilogram bodyweight (p <0.0001) and estimated glomerular filtration rate (eGFR) (p <0.0001) were lower in postmenopausal women. Total daily bolus insulin, daily bolus insulin per kilogram, glycated haemoglobin A1, body mass index and the incidence of severe hypoglycaemic events were similar in both cohorts.Postmenopausal women with type 1 diabetes used lower insulin doses as compared with premenopausal women, whereas glycaemic control and body mass index were comparable. This observation might be explained by lower exposure to sexual hormones and lower eGFR, even though the contribution of other factors such as body composition and eating habits requires further investigation.

Non-High-Density Lipoprotein Cholesterol in Children with Diabetes: Proposed Treatment Recommendations Based on Glycemic Control, Body Mass Index, Age, Sex, and Generally Accepted Cut Points.

Percentile-based non-high-density lipoprotein cholesterol levels were analyzed by glycemic control, weight, age, and sex of children with type 1 diabetes (n = 26,358). Ten percent of all children and 25% of overweight adolescent girls require both immediate lipid-lowering medication and lifestyle changes to achieve non-high-density lipoprotein cholesterol levels <120 mg/dL and cardiovascular risk reduction.

Algorithm-based cholesterol monitoring in children with type 1 diabetes.

OBJECTIVE: To facilitate child-specific and diabetes-related cholesterol control, we developed a monitoring algorithm derived from population-based reference values. STUDY DESIGN: Low-density lipoprotein (LDL)-, non-high-density lipoprotein (HDL)-, and HDL cholesterol percentile values were calculated for children with type 1 diabetes (T1D) and their peers without T1D within algorithm-based categories of sex, age: 1-10 vs >10-<18 years, body mass index: <90th vs ≥90th percentile, and hemoglobin A1c <6%, 6%-<7.5%, 7.5%-9%, >9%. Analyses included 26 147 patients sampled from a German/Austrian population-based registry for T1D (Diabetes Documentation and Quality Management System) and 14 057 peers without diabetes participating in the national Health Interview and Examination Survey for Children and Adolescents in Germany. RESULTS: Reference percentile values for cholesterol were derived as a diagnostic algorithm aimed at supporting long-term cholesterol control. Taking account of a patient's sex, age-group, weight-, and hemoglobin A1c-category, the flowcharts of the algorithm developed separately for LDL-, non-HDL-, and HDL cholesterol allow comparing his/her cholesterol levels with population-based reference percentile values of peers without T1D. CONCLUSIONS: The population-based algorithmic approach applied to LDL-, non-HDL-, and HDL cholesterol allows referencing children with T1D with regard to their peers without T1D and, if necessary, suggests corrections of glycemic control to optimize long-term cholesterol levels.

Direct diabetes-related costs in young patients with early-onset, long-lasting type 1 diabetes.

OBJECTIVE: To estimate diabetes-related direct health care costs in pediatric patients with early-onset type 1 diabetes of long duration in Germany. RESEARCH DESIGN AND METHODS: Data of a population-based cohort of 1,473 subjects with type 1 diabetes onset at 0-4 years of age within the years 1993-1999 were included (mean age 13.9 (SD 2.2) years, mean diabetes duration 10.9 (SD 1.9) years, as of 31.12.2007). Diabetes-related health care services utilized in 2007 were derived from a nationwide prospective documentation system (DPV). Health care utilization was valued in monetary terms based on inpatient and outpatient medical fees and retail prices (perspective of statutory health insurance). Multiple regression models were applied to assess associations between direct diabetes-related health care costs per patient-year and demographic and clinical predictors. RESULTS: Mean direct diabetes-related health care costs per patient-year were €3,745 (inter-quartile range: 1,943-4,881). Costs for glucose self-monitoring were the main cost category (28.5%), followed by costs for continuous subcutaneous insulin infusion (25.0%), diabetes-related hospitalizations (22.1%) and insulin (18.4%). Female gender, pubertal age and poor glycemic control were associated with higher and migration background with lower total costs. CONCLUSIONS: Main cost categories in patients with on average 11 years of diabetes duration were costs for glucose self-monitoring, insulin pump therapy, hospitalization and insulin. Optimization of glycemic control in particular in pubertal age through intensified care with improved diabetes education and tailored insulin regimen, can contribute to the reduction of direct diabetes-related costs in this patient group.

Lipoatrophy is associated with an increased risk of Hashimoto's thyroiditis and coeliac disease in female patients with type 1 diabetes.

BACKGROUND/AIMS: Lipoatrophy (LA) is a rare, possibly under-recognised side effect of insulin treatment of unclear aetiology. The aim of this study was to describe the characteristics of patients with type 1 diabetes (T1D) who have LA and to explore the relationship between LA and other autoimmune diseases based on the hypothesis that additional autoimmune phenomena are more prevalent in T1D patients with LA. METHODS: This was a cross-sectional observational study of T1D patients with LA in comparison to T1D patients without LA who are registered with the Diabetes Patienten-Verlaufsdokumentationssystem database of 241,650 patients in Germany and Austria. RESULTS: Hashimoto's thyroiditis and coeliac disease were more prevalent in patients with LA (p < 0.001 for both). LA was associated with an increased risk of Hashimoto's thyroiditis and coeliac disease in female patients [odds ratio (OR) 2.5, p = 0.003, and OR 3.1, p = 0.02, respectively]. This relationship persisted after adjustment for current age, duration of diabetes and calendar year of treatment (OR 2.7, p = 0.002, and OR 3.5, p = 0.01, respectively). CONCLUSION: These findings support the hypothesis that an immune complex-mediated inflammatory process may be important in the development of LA.

Lower frequency of insulin pump treatment in children and adolescents of Turkish background with type 1 diabetes: analysis of 21,497 patients in Germany.

AIM: This study investigated insulin pump therapy in pediatric patients with type 1 diabetes and Turkish origin compared with those without migration background in Germany. SUBJECTS AND METHODS: Using a nationwide documentation program, we estimated the prevalence of insulin pump therapy in patients < 20 years of age with Turkish origin and those without migration background. Logistic regression was used to adjust for age, sex, diabetes duration, body mass index SD score (BMI-SDS), glycated hemoglobin, number of outpatient visits, number of daily blood glucose self-measurements, and area-based socioeconomic conditions. RESULTS: In 1,695 pediatric type 1 diabetes patients with Turkish background and 19,802 patients without migration background (respectively: 51.2% and 53.0% boys; mean age, 12.4 ± 4.1 and 12.6 ± 4.2 years; mean diabetes duration, 4.7 ± 3.9 and 5.3 ± 4.0 years), fully adjusted prevalences of insulin pump therapy were 18.5% and 30.9%, respectively (odds ratio 0.51, 95% confidence interval 0.43-0.60, P < 0.001). Age, sex, BMI-SDS, outpatient visits, and blood glucose self-control were significantly associated with the prevalence of insulin pump therapy but did not alter the difference substantially. CONCLUSIONS: The prevalence of insulin pump therapy is roughly half among pediatric diabetes patients with Turkish background compared with those without migration background. Several covariates could not explain this difference. Individual characteristics or access barriers within the healthcare system may play a role. Further research is needed.

Improved metabolic control in children and adolescents with type 1 diabetes: a trend analysis using prospective multicenter data from Germany and Austria.

OBJECTIVE: To investigate the temporal trend of metabolic control and potential predictors in German and Austrian children and adolescents with type 1 diabetes. RESEARCH DESIGN AND METHODS: This study is based on a large, multicenter database for prospective longitudinal documentation of diabetes care in Germany and Austria. Data from 30,708 patients documented in 305 diabetes centers between 1995 and 2009 were analyzed. Generalized linear mixed regression models were used to adjust trend analysis for relevant confounders. RESULTS: Unadjusted mean HbA(1c) decreased from 8.7 ± 1.8% in 1995 to 8.1 ± 1.5% in 2009. In multiple regression analysis, treatment year, age, sex, diabetes duration, migration background, BMI-SDS, and daily insulin dose were significant predictors of metabolic control (P < 0.001). After multiple adjustment, mean HbA(1c) decreased significantly by 0.038% per year (95% CI 0.032-0.043%), average odds ratio (OR) per year for HbA(1c) >7.5% (>9.0%) was 0.969 (95% CI 0.961-0.977) (0.948, 95% CI 0.941-0.956). Intensified insulin regimen was associated with lower frequency of poor metabolic control (HbA(1c) >9%; P = 0.005) but not with average HbA(1c) (P = 0.797). Rate of severe hypoglycemia and hypoglycemic coma decreased significantly (relative risk [RR] per year 0.948, 95% CI 0.918-0.979; RR 0.917, 95% CI 0.885-0.950) over the study period. Diabetic ketoacidosis rate showed no significant variation over time. CONCLUSIONS: This study showed a significant improvement in metabolic control in children and adolescents with type 1 diabetes during the past decade and a simultaneous decrease in hypoglycemic events. The improvement was not completely explained by changes in the mode of insulin treatment. Other factors such as improved patient education may have accounted for the observed trend.