Age-period-cohort modelling of type 1 diabetes incidence rates among children included in the EURODIAB 25-year follow-up study.

AIMS: Specific patterns in incidence may reveal environmental explanations for type 1 diabetes incidence. We aimed to study type 1 diabetes incidence in European childhood populations to assess whether an increase could be attributed to either period or cohort effects. METHODS: Nineteen EURODIAB centres provided single year incidence data for ages 0-14 in the 25-year period 1989-2013. Case counts and person years were classified by age, period and cohort (APC) in 1-year classes. APC Poisson regression models of rates were fitted using restricted cubic splines for age, period and cohort per centre and sex. Joint models were fitted for all centres and sexes, to find a parsimonious model. RESULTS: A total of 57,487 cases were included. In ten and seven of the 19 centres the APC models showed evidence of nonlinear cohort effects or period effects, respectively, in one or both sexes and indications of sex-specific age effects. Models showed a positive linear increase ranging from approximately 0.6 to 6.6%/year. Centres with low incidence rates showed the highest overall increase. A final joint model showed incidence peak at age 11.6 and 12.6 for girls and boys, respectively, and the rate-ratio was according to sex below 1 in ages 5-12. CONCLUSION: There was reasonable evidence for similar age-specific type 1 diabetes incidence rates across the EURODIAB population and peaks at a younger age for girls than boys. Cohort effects showed nonlinearity but varied between centres and the model did not contribute convincingly to identification of environmental causes of the increase.

Increased incidence of genital warts among women and men with type 1 diabetes compared with the general population-results from a nationwide registry-based, cohort study.

AIMS: To estimate the incidence rates of genital warts (GWs) in women and men with type 1 diabetes compared to persons without diabetes. METHODS: In this nationwide registry-based cohort study, we included the entire population aged 15 to 49 years living in Denmark between 1996 and 2016. From national registries, we retrieved individual level information on diabetes status, diagnoses and treatment of GWs, and potential confounding variables. We used Poisson regression to model sex- and age-specific incidence rates of GWs in persons with type 1 diabetes and persons without diabetes. Based on the models, we computed sex-specific incidence rate ratios (IRRs) of GWs in persons with type 1 diabetes compared to persons without diabetes, overall and according to age. RESULTS: The analysis included 3,514,824 persons without type 2 diabetes and no GW diagnoses before baseline. The incidence rate of GWs in persons with type 1 diabetes was higher than in those without diabetes, both among women (IRR = 1.59; 95% CI, 1.42-1.78) and men (IRR = 1.36; 95% CI, 1.25-1.48). The pattern of increased incidence rates of GWs in persons with type 1 diabetes was seen at all ages. CONCLUSIONS: Persons with type 1 diabetes have higher incidence rates of GWs than persons without diabetes. This supports the importance of HPV vaccination of young girls and boys with type 1 diabetes.

Effect of metformin on plasma metabolite profile in the Copenhagen Insulin and Metformin Therapy (CIMT) trial.

AIM: Metformin is the first-line treatment for Type 2 diabetes. However, not all people benefit from this drug. Our aim was to investigate the effects of metformin on the plasma metabolome and whether the pretreatment metabolite profile can predict HbA1c outcome. METHODS: Post hoc analysis of the Copenhagen Insulin and Metformin Therapy (CIMT) trial, a multicentre study from May 2008 to December 2012, was carried out. We used a non-target method to analyse 87 plasma metabolites in participants with Type 2 diabetes (n = 370) who were randomized in a 1 : 1 ratio to 18 months of metformin or placebo treatment. Metabolites were measured by liquid chromatography-mass spectrometry at baseline and at 18-month follow-up and the data were analysed using a linear mixed-effect model. RESULTS: At baseline, participants who were on metformin before the trial (n = 312) had higher levels of leucine/isoleucine and five lysophosphatidylethanolamines (LPEs), and lower levels of carnitine and valine compared with metformin-naïve participants (n = 58). At follow-up, participants randomized to metformin (n = 188) had elevated levels of leucine/isoleucine and reduced carnitine, tyrosine and valine compared with placebo (n = 182). At baseline, participants on metformin treatment with the highest levels of carnitine C10:1 and leucine/isoleucine had the lowest HbA1c (P-interaction = 0.02 and 0.03, respectively). This association was not significant with HbA1c at follow-up. CONCLUSIONS: Metformin treatment is associated with decreased levels of valine, tyrosine and carnitine, and increased levels of leucine/isoleucine. None of the identified metabolites can predict the HbA1c -lowering effect of metformin. Further studies of the association between metformin, carnitine and leucine/isoleucine are warranted.

Diabetes-related tuberculosis in Denmark: effect of ethnicity, diabetes duration and year of diagnosis.

BACKGROUND: The association between diabetes mellitus (DM) and tuberculosis (TB) has been established on the basis of cross-sectional studies; however, only a few longitudinal studies have been conducted, with inconsistent results. OBJECTIVE: To study the effect of ethnicity and the presence and duration of DM on the risk of incident TB based on 15 years of follow-up of the entire Danish population. DESIGN AND METHODS: Using Poisson regression analysis, we estimated TB incidence in individuals with DM vs. those without DM by linking nationwide DM and TB registers to the National Civil Register at case level. RESULTS: The TB rate ratio was 1.9 in individuals with DM compared to non-DM individuals, regardless of country of birth, with the exception of African-born individuals (rate ratio 0.5). The risk decreased drastically within the first 2 years after the diagnosis of DM; no association was found with longer durations of DM. The risk also decreased the later the year of DM diagnosis. CONCLUSIONS: The study confirmed DM as a risk factor for TB, except in the case of African-born individuals. Other non-DM risk factors for TB could act as effect-modifiers on the DM-TB association. Implementing earlier DM diagnosis and improving metabolic control may reduce the risk of DM-related TB.

Comparison of Massimo Pronto-7 and HemoCue Hb 201+ with laboratory haemoglobin estimation: a clinical study.

We prospectively studied agreement in haemoglobin estimation between two point-of-care devices (Pronto-7(®) Pulse CO-Oximetry(™), Masimo Corporation, Irvine, California, USA and HemoCue(®) Hb 201 +, HemoCue, Angelholm, Sweden) and an automated laboratory analyser (Sysmex XE5000, Sysmex Corporation, Kobe, Japan). Venous blood sampling and finger co-oximeter readings were performed on 141 pregnant women undergoing routine mid-trimester haemoglobin assessment. Three replicate measures were performed and analysis used Bayesian-based variance component modelling to provide estimates of repeatability, between person within method bias and precision. Repeatability, assessed by coefficient of variation, was higher for Pronto-7(®) (2.3%) compared to HemoCue(®) (5.2%). Fixed bias (mean difference, device - laboratory) was +1.18 (standard deviation 1.19) g/dl and - 0.01 (standard deviation 1.34) g/dl for Pronto-7(®) and HemoCue(®) respectively, with no statistical evidence of proportional bias. Based upon a single device reading, the 95% prediction limits for Pronto-7(®) were -1.2 to 3.6 g.dl-1 and HemoCue(®) were -2.7 to 2.7 g/dl. For both devices precision was not meaningfully improved by averaging replicate readings. However, repeated readings may allow detection of aberrant results. Overall both devices are imprecise and 95% prediction limits wide. We present further prediction limits, derived from the posterior distribution and adjusted for any fixed bias for set levels of probability (certainty). These may be used to support clinical decisions when using these point-of-care devices.

Reproducibility of ultrasonography for assessing abdominal fat distribution in a population at high risk of diabetes.

BACKGROUND: Visceral fat plays an important role in the development of metabolic disease independently of the effect of overall abdominal fat. Ultrasonography is an accessible method of accurately assessing abdominal fat distribution in epidemiological studies, but few details about the reproducibility of this method have been published. OBJECTIVE: The aim of this study was to investigate the reproducibility of ultrasonography in the assessment of abdominal fat distribution in a population at high risk of type 2 diabetes. DESIGN AND METHODS: Ultrasonography was used to estimate visceral and subcutaneous abdominal fat. Intra- and interobserver variation, short-term variation and variation between estimates in the fasting and non-fasting state were examined in three samples of 30, 33 and 23 participants from the ADDITION-PRO study. A variance components model was used to calculate intra- and interobserver variation, and Bland-Altman plots were drawn for all three substudies. RESULTS: Coefficients of variation for intra- and interobserver variation were in the range 3.4-6.1%, except for interobserver variation for subcutaneous fat (9.5%). Short-term variation over a median of 35 days had a coefficient of variation of 15%. The effect of a meal was primarily on the visceral estimates and did not extend beyond the first postprandial hour. Non-fasting visceral estimates were larger than fasting estimates. CONCLUSION: Both visceral and subcutaneous fat can be estimated with ultrasonography with adequate intra- and interobserver reproducibility by clinical researchers with limited training, making it a feasible method of assessing abdominal fat distribution in epidemiological studies.

Diabetes and cancer (1): evaluating the temporal relationship between type 2 diabetes and cancer incidence.

Substantial evidence suggests that people with type 2 diabetes have an increased risk of developing several types of cancers. These associations may be due to a number of direct and indirect mechanisms. Observational studies of these associations, including the potential role for glucose-lowering therapy, are being increasingly reported, but face a number of methodological challenges. This paper is the first of two review papers addressing methodological aspects underpinning the interpretations of links between diabetes and cancer, and suggests potential approaches to study designs to be considered in observational studies. This paper reviews factors related to cancer incidence in the diabetic population; the second paper relates to studies of cancer mortality.

Causes of death among diabetic patients in Denmark.

AIMS/HYPOTHESIS: To study the distribution of causes of death in the Danish population, and its variation by diabetes status, sex, age and calendar year as well as the years of life lost from the specific causes of death. METHODS: Persons aged 30-98 years were followed from 1995 to 2008 by linkage of Danish registers. Poisson regression was used to model cause-specific mortality rates by age and calendar time for each specific cause of death, according to sex and diabetes status. The mortality rates were also modelled as a function of age and birth cohort. We computed the distribution of causes of death and years of life lost from specific causes of death due to diabetes. RESULTS: During the 14-year study period, patients with diabetes contributed 2.3 million person-years of follow-up and 124,210 deaths, whereas persons without diabetes contributed 45.1 million person-years and 648,020 deaths. The mortality was higher among individuals with diabetes, and the mortality ratio (diabetes vs no diabetes) decreased with age and for all causes and cardiovascular diseases also by calendar time. The effect of sex on the association between diabetes and mortality varied with age and cause of death. About 9 years of life were lost to diabetes at age 30 years, and 3 years at age 70 years. CONCLUSIONS/INTERPRETATION: Age-specific mortality is higher among people with diabetes, and rate ratios vary with age, sex, calendar period and cause of death. The distribution of causes of death was similar for persons with and without diabetes.

Cancer occurrence in Danish diabetic patients: duration and insulin effects.

AIMS/HYPOTHESIS: Cancer is more frequent among diabetes patients, but it is unknown how this excess varies with duration of diabetes and insulin use. The aim of this study was to analyse disease data to examine this issue further. METHODS: We linked the Danish National Diabetes Register and Cancer Registry and performed a cohort analysis of the entire Danish population by diabetes status, duration of diabetes and insulin use, comparing cancer incidence rates in diabetic patients with the non-diabetic population for the 15 year period 1995-2009, using Poisson regression with natural splines to describe the variation by duration. RESULTS: We found 20,032 cancer cases among patients not using insulin and 2,794 cancer cases among diabetic patients using insulin. The cancer incidence rate ratio among non-insulin users relative to the non-diabetic population decreased from over 2 at diagnosis to 1.15 after 2 years of diabetes duration. The cancer incidence rate ratio was higher among patients using insulin, decreasing from 5 at the start of insulin treatment to about 1.3 [corrected] after 5 years of insulin use. Among non-insulin users, cancers of the stomach, colorectum, liver, pancreas, lung, corpus uteri, kidney and brain, and lymphomas were elevated. Among insulin users the rate ratio of prostate cancer was decreasing by duration whereas we found higher risk of cancer of the stomach, lung, liver, pancreas and kidney. Breast cancer incidence rates were not affected by either diabetes or insulin use. CONCLUSIONS: The observed duration effects suggest that both increased surveillance for cancer in the first years after diagnosis of diabetes, and reverse causation, where undiagnosed cancers increase the likelihood of diabetes diagnosis, play a role. For longer durations, a combination of common causes for diabetes and cancer, as well as the effects of diabetes and insulin exposure per se, may play a role in the association between diabetes and some cancers.

HbA1(c) and mean blood glucose show stronger associations with cardiovascular disease risk factors than do postprandial glycaemia or glucose variability in persons with diabetes: the A1C-Derived Average Glucose (ADAG) study.

AIMS: Increased glucose excursions and postprandial hyperglycaemia have been suggested as unique risk factors for cardiovascular disease (CVD) and mortality in patients with diabetes mellitus. Much of the evidence is based on a single 2 h glucose value after oral glucose tolerance testing in epidemiological studies. We examined the association between various indices of glycaemia measured during everyday activities and metabolic CVD risk factors in the A1C-Derived Average Glucose (ADAG) study. METHODS: Participants (268 with type 1 diabetes, 159 with type 2 diabetes) completed 16 weeks of intensive continuous glucose monitoring (CGM) and self-monitoring of blood glucose (SMBG). From these data, common indices of postprandial glycaemia, overall hyperglycaemia, glucose variability and HbA1(c) were derived. The associations between glycaemic indices and known CVD risk factors (lipids, high-sensitivity C-reactive protein and blood pressure) were explored in linear regression models. RESULTS: For both diabetes types, the overall strongest associations with CVD risk factors were seen for the measures of average glycaemia (mean blood glucose and HbA1(c)). Associations between self-monitored postprandial and fasting glucose and CVD risk factors were weaker, but significant. Measurements of blood glucose variability showed non-significant associations. Overall, calculations based on CGM were not more informative than those based on frequent SMBG. CONCLUSIONS/INTERPRETATION: Mean glycaemia and HbA1(c) show consistent and stronger associations with CVD risk factors than fasting glucose or postprandial glucose levels or measures of glucose variability in patients with diabetes.

Low birthweight and premature birth are both associated with type 2 diabetes in a random sample of middle-aged Danes.

AIMS/HYPOTHESIS: We studied the associations of size at birth and prematurity with type 2 diabetes, insulin sensitivity and beta cell function in the Danish population-based Inter99 study (ClinicalTrials.gov NCT00289237). METHODS: Information about size at birth and prematurity was identified from original midwife records in 4,744 middle-aged Danes. Type 2 diabetes status, insulin sensitivity (Matsuda index) and beta cell function (disposition index) were assessed using a 75 g oral glucose tolerance test. Participants born prematurely were compared with a group of at-term participants born small for gestational age. RESULTS: An increase in birthweight of 1 kg was associated with a 51% (OR 0.49, 95% CI 0.35-0.69) reduced risk of type 2 diabetes. Ponderal index, reflecting thinness at birth, was associated with type 2 diabetes to the same extent as birthweight. The prevalence of type 2 diabetes was increased to a similar degree in participants born prematurely and participants born small for gestational age, although the former had a higher ponderal index at birth. In addition, birthweight z-scores, reflecting fetal growth rate, were unrelated to the risk of type 2 diabetes and to other measures of glucose regulation in participants born prematurely. While low birthweight was inversely associated with insulin sensitivity and beta cell function, prematurity was associated solely with decreased insulin sensitivity. CONCLUSIONS/INTERPRETATION: While the association between birthweight and risk of type 2 diabetes is mediated via combined effects on beta cell function and insulin sensitivity, prematurity seems to influence risk of type 2 diabetes via attenuated insulin sensitivity only and independently of fetal growth rates.

Real-life glycaemic profiles in non-diabetic individuals with low fasting glucose and normal HbA1c: the A1C-Derived Average Glucose (ADAG) study.

AIMS/HYPOTHESIS: Real-life glycaemic profiles of healthy individuals are poorly studied. Our aim was to analyse to what extent individuals without diabetes exceed OGTT thresholds for impaired glucose tolerance (IGT) and diabetes. METHODS: In the A1C-Derived Average Glucose (ADAG) study, 80 participants without diabetes completed an intensive glucose monitoring period of 12 weeks. From these data, we calculated the average 24 h glucose exposure as time spent above different plasma glucose thresholds. We also derived indices of postprandial glucose levels, glucose variability and HbA(1c). RESULTS: We found that 93% of participants reached glucose concentrations above the IGT threshold of 7.8 mmol/l and spent a median of 26 min/day above this level during continuous glucose monitoring. Eight individuals (10%) spent more than 2 h in the IGT range. They had higher HbA(1c), fasting plasma glucose (FPG), age and BMI than those who did not. Seven participants (9%) reached glucose concentrations above 11.1 mmol/l during monitoring. CONCLUSIONS/INTERPRETATION: Even though the non-diabetic individuals monitored in the ADAG study were selected on the basis of a very low level of baseline FPG, 10% of these spent a considerable amount of time at glucose levels considered to be 'prediabetic' or indicating IGT. This highlights the fact that exposure to moderately elevated glucose levels remains under-appreciated when individuals are classified on the basis of isolated glucose measurements.

The Danish National Diabetes Register: trends in incidence, prevalence and mortality.

AIMS/HYPOTHESIS: The aim of the study was to describe trends in the incidence rate, prevalence and mortality rate for diabetes in Denmark. METHODS: Healthcare registers at the National Board of Health were used to compile a register of diabetic patients in the Danish population (5.4 million people). Age- and sex-specific prevalence, incidence rates, mortality rates and standardised mortality ratios relative to the non-diabetic part of the population were calculated. RESULTS: The register contains records for about 360,000 persons with diabetes; 230,000 were alive at 1 January 2007, corresponding to an overall prevalence of 4.2%. The prevalence increased by 6% per year. In 2004 the incidence rates were 1.8 per 100,000 at age 40 years and 10.0 per 100,000 at age 70 years. The incidence rate increased 5% per year before 2004 and then stabilised. The mortality rate in the diabetic population decreased 4% per year, compared with 2% per year in the non-diabetic part of the population. The mortality rate decreased 40% during the first 3 years after inclusion in the register. The standardised mortality ratio decreased with age, from 4.0 at age 50 years to 2.5 at age 70 years and just under 2 at age 85 years, identically for men and women. The standardised mortality ratio decreased 1% per calendar year. The lifetime risk of diabetes was 30%. CONCLUSIONS/INTERPRETATION: The prevalence of diabetes in Denmark rose in 1995-2006, but the mortality rate in diabetic patients decreased faster than that of the non-diabetic population. The mortality rate decreased markedly just after inclusion in the register. Incidence rates have shown a tendency to decrease during the last few years, but this finding should be viewed with caution.

Short-term reproducibility of impaired fasting glycaemia, impaired glucose tolerance and diabetes The ADDITION study, DK.

We evaluated variations in glucose measurements and the reproducibility of glucose tolerance classification in a high-risk screening setting in general practice. Screening for diabetes was performed in persons aged 40-69 years. Based on capillary fasting (FBG) and 2-h blood glucose (2 hBG) individuals with impaired fasting glycaemia (IFG), impaired glucose tolerance (IGT) and diabetes had a second test done after 14 days. Intra-individual coefficients of variation (CV) were estimated in each glucose tolerance class using the approximation CV(2)(x)=var(ln(x)). Bland-Altman plots with limits of agreement were made. In the total population, the CV(intra) was 7.9% and 13.8% for FBG and 2 hBG, respectively. Limits of agreement ranged from -1.15 to 1.67 mmol/l for FBG and from - 2.62 to 3.27 mmol/l for 2 hBG. One individual with IFG and 22.5% with IGT had diabetes at the second test, 76.1% with diabetes had this diagnosis confirmed, and about 30% with IFG and IGT had normal glucose tolerance at the second test. The expected values of repeated capillary blood glucose measurements were about+/-1 and+/-3 mmol/l for FBG and 2 hBG, respectively. Yet, 70% of high-risk prediabetic individuals were persistently classified with abnormal glucose regulation; diabetes was confirmed in 76% of the cases.

Seasonality of birth in children and young adults (0-29 years) with type 1 diabetes in Ukraine.

AIMS/HYPOTHESIS: Numerous epidemiological studies have shown differences in seasonality of birth patterns between the general population and the group who develop type 1 diabetes mellitus. This finding indicates that environmental factors operating during pre- and/or postnatal development could be aetiologically important. We examined whether the pattern of month of birth for type 1 diabetes patients in Ukraine differs from that for total live births. METHODS: Data consist of prevalent cases of type 1 diabetes in Ukraine by the end of 2003. Cases are restricted to persons born after 1 January 1960, diagnosed with type 1 diabetes before the age of 30 years (n = 20,117). People born during the same time in the general population (n = 29,105,560) were the reference standard. Seasonal patterns were estimated using logistic regression with harmonic terms. RESULTS: We found a strongly significant seasonal pattern of type 1 diabetes incidence rates (p < 0.001), with the lowest rates in December and the highest in April. The rate ratio between the extremes was 1.32 (95% CI 1.27-1.39). Tests for seasonal patterns in subgroups defined by sex and age or by sex and date of birth were all significant with p values less than 0.02. We found no interactions with sex (p = 0.142) or age at diagnosis (p = 0.207), but found a strong interaction with period of birth (p < 0.0001). CONCLUSIONS/INTERPRETATION: The results obtained indicate that early-life factors linked to seasons may influence type 1 diabetes risk later in life.

Age-period-cohort models for the Lexis diagram.

Analysis of rates from disease registers are often reported inadequately because of too coarse tabulation of data and because of confusion about the mechanics of the age-period-cohort model used for analysis. Rates should be considered as observations in a Lexis diagram, and tabulation a necessary reduction of data, which should be as small as possible, and age, period and cohort should be treated as continuous variables. Reporting should include the absolute level of the rates as part of the age-effects. This paper gives a guide to analysis of rates from a Lexis diagram by the age-period-cohort model. Three aspects are considered separately: (1) tabulation of cases and person-years; (2) modelling of age, period and cohort effects; and (3) parametrization and reporting of the estimated effects. It is argued that most of the confusion in the literature comes from failure to make a clear distinction between these three aspects. A set of recommendations for the practitioner is given and a package for R that implements the recommendations is introduced.

Frequency and risk factors of severe hypoglycaemia in insulin-treated Type 2 diabetes: a cross-sectional survey.

AIMS: The reported risk of severe hypoglycaemia in insulin-treated Type 2 diabetes is highly variable and few studies have evaluated the influence of risk factors. We assessed the incidence and the influence of potential risk factors for severe hypoglycaemia in a questionnaire survey in subjects with insulin-treated Type 2 diabetes receiving currently recommended multifactorial intervention. METHODS: Consecutive patients with insulin-treated Type 2 diabetes (n = 401) completed a questionnaire about occurrence of hypoglycaemia in the past, hypoglycaemia awareness and socio-demographic factors. A zero-inflated negative binomial model was used to assess the influence of potential risk factors on the rate of severe hypoglycaemia. RESULTS: The overall incidence of severe hypoglycaemia in the preceding year was 0.44 episodes/person year. Sixty-six (16.5%) patients had experienced at least one event. The risk of any episode of severe hypoglycaemia positively correlated with impaired hypoglycaemia awareness, being married and long duration of diabetes. The risk of repeated episodes of severe hypoglycaemia positively correlated with the presence of peripheral neuropathy, while long duration of diabetes prior to insulin treatment and treatment with angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor antagonists (ARBs) were associated with reduced risk. C-peptide concentration and HbA1c were not associated with the risk of severe hypoglycaemia. CONCLUSIONS: In this cohort of insulin-treated Type 2 diabetic patients, the incidence of severe hypoglycaemia is higher than reported in most studies, corresponding to about one-third of that in Type 1 diabetes. Impaired hypoglycaemia awareness is the most important risk factor for severe hypoglycaemia.