Preferences for treatment among adolescents with Type 1 diabetes: a national study using a discrete choice experiment model.

AIM: To test the possibility of using a discrete choice experiment model, on a national level in adolescents with Type 1 diabetes, in order to obtain a better understanding of drivers of and barriers to diabetes self-care. METHODS: A survey instrument was constructed and tested on a small group of the target population: adolescents aged 15 to <18 years with Type 1 diabetes. All individuals in Sweden belonging to this target group (N=2112) were then identified via the Swedish paediatric diabetes quality registry SWEDIABKIDS, and were sent an invitation to answer an online questionnaire. A valid response for the discrete choice experiment analyses was achieved from 431 individuals. RESULTS: The included respondents were not statistically different from non-participants in terms of age and duration of diabetes, but more young women entered the study and the participants had (on average) a significantly lower HbA1c value than the non-participants. Participants regarded as undesirable both non-severe hypoglycaemic events (day and night) and hyperglycaemic events. Avoiding weight gain and even achieving weight loss were the most important aspects among female respondents, who were willing to trade off a substantial level of glycaemic control [13 mmol/mol (1.2%)] to avoid a weight gain of 3 kg. Hypothetical equipment improvements were desired. CONCLUSIONS: The responses may provide useful indications of the aspects that the respondents would prioritize given a real-life dilemma. For treatment effects, stratification along gender lines was important, whereas the treatment administration aspects were stratified according to treatment type because these aspects are closely related.

Body mass index standard deviation score and obesity in children with type 1 diabetes in the Nordic countries. HbA1c and other predictors of increasing BMISDS.

BACKGROUND: Intensified insulin therapy may increase body weight and cause obesity. This study compared body mass index standard deviation score (BMISDS) and obesity rate in children with type 1 diabetes (T1D) in Denmark, Iceland, Norway and Sweden, and uncovered predictors for increasing BMISDS. METHODS: Data registered in the Nordic national childhood diabetes databases during the period 2008-2012 on children below 15 years with T1D for more than 3 months were compiled, including information on gender, age, diabetes duration, hemoglobin A1c (HbA1c ), insulin dose, severe hypoglycemia (SH), treatment modality, height and weight. The Swedish reference chart for BMI was used for calculating BMISDS. RESULTS: Totally, 11 025 children (48% females) (30 994 registrations) were included. Medians by the last recorded examination were: age, 13.5 years; diabetes duration, 4.3 years; HbA1c , 7.9% (63 mmol/mol); insulin dose, 0.8 IU/kg/d and BMISDS, 0.70. Obesity rate was 18.5%. Adjusted mean BMISDS (BMISDS adj) was inversely related to HbA1c and directly to diabetes duration. Higher BMISDS adj was found in those with an insulin dose above 0.6 IU/kg/d, and in girls above 10 years. Pump users had higher BMISDS adj than pen users, and patients with registered SH had higher BMISDS adj than patients without SH (both P < .001). CONCLUSION: Obesity rate in children with T1D in the Nordic countries is high, however, with country differences. Low HbA1c , long diabetes duration, higher insulin dose, pump treatment and experiencing a SH predicted higher BMISDS. Diabetes caregivers should balance the risk of obesity and the benefit of a very low HbA1c.

Antibodies to influenza virus A/H1N1 hemagglutinin in Type 1 diabetes children diagnosed before, during and after the SWEDISH A(H1N1)pdm09 vaccination campaign 2009-2010.

We determined A/H1N1-hemagglutinin (HA) antibodies in relation to HLA-DQ genotypes and islet autoantibodies at clinical diagnosis in 1141 incident 0.7-to 18-year-old type 1 diabetes patients diagnosed April 2009-December 2010. Antibodies to (35) S-methionine-labelled A/H1N1 hemagglutinin were determined in a radiobinding assay in patients diagnosed before (n = 325), during (n = 355) and after (n = 461) the October 2009-March 2010 Swedish A(H1N1)pdm09 vaccination campaign, along with HLA-DQ genotypes and autoantibodies against GAD, insulin, IA-2 and ZnT8 transporter. Before vaccination, 0.6% patients had A/H1N1-HA antibodies compared with 40% during and 27% after vaccination (P < 0.0001). In children <3 years of age, A/H1N1-HA antibodies were found only during vaccination. The frequency of A/H1N1-HA antibodies during vaccination decreased after vaccination among the 3 < 6 (P = 0.006) and 13 < 18 (P = 0.001), but not among the 6 < 13-year-olds. HLA-DQ2/8 positive children <3 years decreased from 54% (15/28) before and 68% (19/28) during, to 30% (9/30) after vaccination (P = 0.014). Regardless of age, DQ2/2; 2/X (n = 177) patients had lower frequency (P = 0.020) and levels (P = 0.042) of A/H1N1-HA antibodies compared with non-DQ2/2; 2/X (n = 964) patients. GADA frequency was 50% before, 60% during and 51% after vaccination (P = 0.009). ZnT8QA frequency increased from 30% before to 34% during and 41% after vaccination (P = 0.002). Our findings suggest that young (<3 years) along with DQ2/2; 2/X patients were low responders to Pandemrix(®) . As the proportion of DQ2/8 patients <3 years of age decreased after vaccination and the frequencies of GADA and ZnT8QA were enhanced, it cannot be excluded that the vaccine affected clinical onset of type 1 diabetes.

Impaired metabolic control and socio-demographic status in immigrant children at onset of Type 1 diabetes.

AIM: The aim of the present study was to compare clinical and socio-demographic conditions at the onset of Type 1 diabetes in children born to immigrant families and children born to Swedish families, and to assess whether those conditions had an impact on metabolic status. METHODS AND DESIGN: This was an observational nationwide population-based matched cohort study on prospectively recorded registry data of all children with diabetes in Sweden and their families during 2000-2010. Out of a total of 13 415 children from the Swedish Childhood Diabetes Registry (SWEDIABKIDS), 879 children born to immigrant parents were collected. To these we added 2627 children with Swedish-born parents, matched for gender, age and year of onset of Type 1 diabetes. RESULTS: The proportion of low capillary pH (< 7.30) at onset was higher in the immigrant cohort [25.8% vs. 16.4% in the Swedish cohort (P < 0.001)]. HbA1c was also higher [95 mmol/mol (10.8%) vs. 88 mmol/mol (10.2%), respectively (P < 0.001)]. In a logistic regression model with low pH as the dependent variable, we were unable to reveal any significant association to socio-demographic factors, but the odds ratio for HbA1c was 0.983 (95% CI 0.976-0.991) and for plasma glucose was 0.953 (95% CI 0.933-0.973). CONCLUSION: Children born to immigrant parents have lower capillary pH and higher HbA1c at diabetes onset. Immigrant families harbour lower socio-demographic living conditions, but this fact does not seem to influence the inferior metabolic condition at diabetes onset.

Glycated haemoglobin variations in paediatric type 1 diabetes: the impact of season, gender and age.

AIM: To study whether monthly variations in type 1 diabetes incidence are related to monthly glycated haemoglobin (HbA1c) levels at diagnosis and if high HbA1c at diagnosis is related to certain clinical variables at diagnosis and during the clinical course of the disease. METHODS: Data from 4430 boys and 3590 girls registered in the Swedish paediatric diabetes quality registry, Swedish paediatric diabetes quality registry, from 2000 to 2010 were analysed. RESULTS: Month of onset varied (p < 0.001), with 53% diagnosed during September to February, and mean HbA1c at diagnosis was highest in May (10.9%, 96 mmol/mol) and lowest in (October 9.4%, 88 mmol/mol) (p < 0.001). Girls showed higher HbA1c at onset than boys (p < 0.001). More than half (53%) with an annual mean HbA1c of >9.3% (78 mmol/mol) and 4% of those with an annual mean of <7.4% (57 mmol/mol) in 2007 had >9.3% (78 mmol/mol) in 2010. CONCLUSION: Patients with high HbA1c levels during a certain period have the same high levels several years later. This group, perhaps including those with high HbA1c level at diagnosis, may need more intensive care, including extra support from the diabetes teams and other forms of medical treatment.

Thyroid autoimmunity in relation to islet autoantibodies and HLA-DQ genotype in newly diagnosed type 1 diabetes in children and adolescents.

AIMS/HYPOTHESIS: The aim of this work was to investigate, in children newly diagnosed with type 1 diabetes: (1) the prevalence of autoantibodies against thyroid peroxidase (TPOAb) and thyroglobulin (TGAb); and (2) the association between TPOAb, TGAb or both, with either islet autoantibodies or HLA-DQ genes. METHODS: Blood samples from 2,433 children newly diagnosed with type 1 diabetes were analysed for TPOAb and TGAb in addition to autoantibodies against arginine zinc transporter 8 (ZnT8RA), tryptophan zinc transporter 8 (ZnT8WA), glutamine zinc transporter 8 (ZnT8QA), glutamic acid decarboxylase (GADA), insulin (IAA), insulinoma-associated protein-2 (IA-2A), HLA-DQA-B1 genotypes, thyroid-stimulating hormone (TSH) and free thyroxine (T4). RESULTS: At type 1 diabetes diagnosis, 12% of the children had thyroid autoantibodies (60% were girls; p < 0.0001). GADA was positively associated with TPOAb (p < 0.001) and with TGAb (p < 0.001). In addition, ZnT8A was associated with both TPOAb (p = 0.039) and TGAb (p = 0.015). DQB1*05:01 in any genotype was negatively associated with TPOAb (OR 0.55, 95% CI 0.37, 0.83, p value corrected for multiple comparisons (p c) = 0.012) and possibly with TGAb (OR 0.55, 95% CI 0.35, 0.87, p c = 0.07). Thyroid autoimmunity in children newly diagnosed with type 1 diabetes was rarely (0.45%) associated with onset of clinical thyroid disease based on TSH and free T4. CONCLUSIONS/INTERPRETATION: GADA and ZnT8A increased the risk for thyroid autoimmunity at the time of clinical diagnosis of type 1 diabetes, while HLA-DQB1*05:01 reduced the risk. However, the associations between thyroid autoimmunity and HLA-DQ genotype were weak and did not fully explain the co-occurrence of islet and thyroid autoimmunity.

Residual beta cell function at diagnosis of type 1 diabetes in children and adolescents varies with gender and season.

BACKGROUND: There are seasonal variations and gender differences in incidence of type 1 diabetes (T1D), metabolic control and responses to immune interventions at onset of the disease. We hypothesized that there are seasonal and gender differences in residual insulin secretion already at diagnosis of T1D. METHODS: In 2005, a national study, the Better Diabetes Diagnosis, was started to classify all newly diagnosed children and adolescents with diabetes. About 95% (3824/4017) of the patients were classified as T1D, and our analyses are based on the patients with T1D. RESULTS: C-peptide was lower in younger children, 0-10 years of age (0.23 ± 0.20 nmol/L) than in older children, 11-18 years of age (0.34 ± 0.28 nmol/L) (p < 0.000 ). There was a seasonal variation in non-fasting serum C-peptide, significantly correlated to the seasonal variation of diagnosis (p < 0.01). Most children were diagnosed in January, February and March as well as in October when C-peptide was highest, whereas fewer patients were diagnosed in April and May when serum C-peptide was significantly lower (p < 0.01). The seasonal variation of C-peptide was more pronounced in boys than in girls (p < 0.000 and p < 0.01, respectively). Girls had higher C-peptide than boys (p < 0.05), especially in early puberty. CONCLUSIONS: Both seasonal and gender differences in residual beta cell function exist already at diagnosis of T1D. These observations have consequences for treatment and for randomizing patients in immune intervention clinical trials.

Triple specificity of ZnT8 autoantibodies in relation to HLA and other islet autoantibodies in childhood and adolescent type 1 diabetes.

OBJECTIVE: To establish the diagnostic sensitivity of and the relationships between autoantibodies to all three Zinc transporter 8 (Zinc transporter 8 autoantibody to either one, two, or all three amino acid variants at position 325, ZnT8A) variants to human leukocyte antigen (HLA)-DQ and to autoantibodies to glutamic acid decarboxylase (GADA), insulinoma-associated protein 2 (IA-2A), and insulin (IAA). METHODS: We analyzed 3165 patients with type 1 diabetes (T1D) in the Better Diabetes Diagnosis study for HLA-DQ genotypes and all six autoantibodies (ZnT8RA, arginine 325 Zinc transporter 8 autoantibody; ZnT8WA, tryptophan 325 Zinc transporter 8 autoantibody; ZnT8QA, glutamine 325 Zinc transporter 8 autoantibody; GADA, IA-2A, and IAA). RESULTS: ZnT8A was found in 65% of the patients and as many as 108 of 3165 (3.4%) had 1-3 ZnT8A alone. None had ZnT8QA alone. Together with GADA (56%), IA-2A (73%), and IAA (33%), 93% of the T1D patients were autoantibody positive. All three ZnT8A were less frequent in children below 2 yr of age (p < 0.0001). All three ZnT8A were associated with DQA1-B1*X-0604 (DQ6.4) and DQA1-B1*03-0302 (DQ8). ZnT8WA and ZnT8QA were negatively associated with DQA1-B1*05-02 (DQ2). CONCLUSIONS: Analysis of ZnT8A increased the diagnostic sensitivity of islet autoantibodies for T1D as only 7% remained islet autoantibody negative. The association between DQ6.4 and all three ZnT8A may be related to ZnT8 antigen presentation by the DQ6.4 heterodimer.

C-peptide in the classification of diabetes in children and adolescents.

AIM: To report C-peptide results in newly diagnosed patients and the relation to clinical diagnosis of diabetes. METHODS: A nation-wide cohort, the Better Diabetes Diagnosis study was used to determine serum C-peptide at diagnosis in 2734 children and adolescents. Clinical data were collected at diagnosis and follow-up. C-peptide was determined in a validated and controlled time-resolved fluoroimmunoassay. RESULTS: The clinical classification of diabetes, before any information on human leukocyte antigen, islet autoantibodies, or C-peptide was received, was type 1 diabetes (T1D) in 93%, type 2 diabetes (T2D) in 1.9%, maturity onset diabetes of the young (MODY) in 0.8%, secondary diabetes (0.6%), while 3.3% could not be classified. In a random, non-fasting serum sample at diagnosis, 56% of the patients had a C-peptide value >0.2 nmol/L. Children classified as T2D had the highest mean C-peptide (1.83 + 1.23 nmol/L) followed by MODY (1.04 ± 0.71 nmol/L) and T1D (0.28 ± 0.25 nmol/L). Only 1/1037 children who had C-peptide <0.2 nmol/L at diagnosis was classified with a type of diabetes other than T1D. Predictive value of C-peptide >1.0 nmol/L for the classification of either T2D or MODY was 0.46 [confidence interval 0.37-0.58]. CONCLUSIONS: More than half of children with newly diagnosed diabetes have clinically important residual beta-cell function. As the clinical diagnosis is not always straightforward, a random C-peptide taken at diagnosis may help to classify diabetes. There is an obvious use for C-peptide determinations to evaluate beta-cell function in children with diabetes.

A low incidence of Type 1 diabetes between 1977 and 2001 in south-eastern Sweden in areas with high population density and which are more deprived.

AIMS: To explore how socioeconomic factors and population density may contribute to the geographical variation of incidence of Type 1 diabetes in children in south-eastern Sweden. METHOD: All children diagnosed with Type 1 diabetes in south-eastern Sweden during 1977-2001 were defined geographically to their place of residence and were allocated x and y coordinates in the national grid. The population at risk and socioeconomic data were aggregated in 82,000 200-m squares and geocoded likewise. A socioeconomic index was calculated using a signed chi(2) method. Rural-urban gradients were defined by overlay analysis in a geographic information system. RESULTS: The incidence during the past 25 years has been rising steadily, particularly in the last 6 years. The incidence was highest in areas with a high proportion of small families, of families with a high family income and better education, and this was found both at the time of diagnosis and at the time of birth. In the rural-urban analysis, the lowest incidence was found in the urban area with > 20,000 inhabitants, where there was also a higher frequency of deprivation. CONCLUSIONS: Our findings indicate that geographical variations in incidence rates of Type 1 diabetes in children are associated with socioeconomic factors and population density, although other contributing factors remain to be explained.

The Better Diabetes Diagnosis (BDD) study - A review of a nationwide prospective cohort study in Sweden.

The incidence of type 1 diabetes (T1D) in Sweden is one of the highest in the world. However, the possibility of other types of diabetes must also be considered. In addition, individuals with T1D constitute a heterogeneous group. A precise classification of diabetes is a prerequisite for optimal outcome. For precise classification, knowledge on the distribution of genetic factors, biochemical markers and clinical features in individuals with new onset of diabetes is needed. The Better Diabetes Diagnosis (BDD), is a nationwide study in Sweden with the primary aim to facilitate a more precise classification and diagnosis of diabetes in order to enable the most adequate treatment for each patient. Secondary aims include identification of risk factors for diabetes-related co-morbidities. Since 2005, data on almost all children and adolescents with newly diagnosed diabetes in Sweden are prospectively collected and including heredity of diabetes, clinical symptoms, levels of C peptide, genetic analyses and detection of autoantibodies. Since 2011, analyses of HLA profile, autoantibodies and C peptide levels are part of clinical routine in Sweden for all pediatric patients with suspected diagnosis of diabetes. In this review, we present the methods and main results of the BDD study so far and discuss future aspects.

Seasonal variation in the diagnosis of type 1 diabetes in south-east Sweden.

With the aim to survey the seasonal pattern of diagnosis of type 1 diabetes we included all 1903 children <16 years of age and who had been diagnosed with type 1 diabetes between 1977 and 2001 in the south-east of Sweden. To investigate the seasonal pattern a mixture of two cosine functions was included in a logistic regression model. There was a clear seasonal variation over the years (p<0.001). Children in the oldest age group (11-15 years) showed the most obvious seasonal variation (p<0.001). Children with a short duration of symptoms had about the same seasonal variation as children with a long duration. Both children with and without an infection 3 months prior to diagnosis showed significant seasonal variation (p<0.001) although the seasonal pattern differed between the two groups (p<0.001). As the incidence of diabetes increased during the 25 years the study period was divided into periods of 5 years and it was only during the two last periods that significant seasonal variation occurred. There is a clear seasonal variation in diagnosis of type 1 diagnosis in children and the results suggest that children with a less aggressive disease process at diagnosis were most responsible for this variation. Children with and without prior infection showed a different seasonal pattern.

Incidence of severe hypoglycemia in children with type 1 diabetes in the Nordic countries in the period 2008-2012: association with hemoglobin A 1c and treatment modality.

OBJECTIVE: Treatment of type 1 diabetes has been intensified aiming at normalizing blood glucose, which may increase the risk of severe hypoglycemia (SH). We aimed to compare the incidence of SH events in the four Nordic countries Denmark, Iceland, Norway and Sweden, and to assess the influence of hemoglobin A1c (HbA1c) and treatment modalities on the frequency of SH; particularly, to explore if a HbA1c target ≤6.7% (50 mmol/mol) is feasible. RESEARCH DESIGN AND METHODS: Data on children below 15 years with a diabetes duration more than 1 year, registered in the national childhood diabetes databases in the four Nordic countries from 2008 to 2012, were compiled. Data completeness was more than 95%. RESULTS: Totally 8806 (48% females) patients with 29 715 person years were included, mean age and diabetes duration were 11 years and 5.1 years, respectively. The overall rate of SH was 6.0 per 100 patient-years, and did not change during the study period. The Swedish population constantly had the lowest SH incidence while it decreased significantly in the Danish population. HbA1c decreased significantly over time (p<0.01), while the number of pump users increased (p<0.01). Stratifying for HbA1c levels showed the lowest risk of SH in patients with HbA1c ≤6.7% (≤50 mmol/mol), but in the statistical models adjusting for possible confounders the difference between the HbA1c groups disappeared. Pump users had the lowest SH risk, also after adjusting for possible confounders. CONCLUSIONS: Risk of SH differs between the Nordic countries with the lowest risk in Sweden. Pump therapy was associated with decreased risk of SH. The low HbA1c group had the same or a lower risk of SH compared with the highest HbA1c groups. A target HbA1c ≤6.7% (≤50 mmol/mol) seems achievable without increasing the risk of SH.

Space-time clustering in insulin-dependent diabetes mellitus (IDDM) in south-east Sweden.

Using the method developed by Knox, space-time clustering was analysed in all 584 cases of insulin-dependent diabetes mellitus (IDDM) diagnosed between 1977-1990 and below the age of 16 from four paediatric departments in south-east Sweden. The catchment areas of these clinics form a contiguous geographical area of 26,445 km2 with a total of 165,784 children aged 0-15 years. The annual incidence for the study period was 25.2 per 100,000 children aged 0-15 years with a statistically significant incidence variation between the years; highest incidence in 1983 (39.2) and lowest in 1977 and 1989 (18.9 and 20.7 respectively). In the analysis, 30 different combinations of critical cutoff values were used to define closeness in space and time of pairs of cases. Statistically significant results were seen for several combinations with the highest significance obtained for the cutoff values of 15 km and 7 months, respectively. This space-time clustering tended to be specially pronounced during the period with the highest incidence of IDDM, 1981-1985. Our results indicate that space-time clustering exists in IDDM. These findings may support the hypothesis that infectious agents (possibly viral) give rise to a portion of IDDM.

Geographical mapping of type 1 diabetes in children and adolescents in south east Sweden.

STUDY OBJECTIVE: As earlier studies have shown space-time clusters at onset of type 1 diabetes in the south east region of Sweden we investigated if there also has been any geographical clusters of diabetes in this region. DESIGN: The place of residence (coordinates) at the time of diagnosis were geocoded in a geographical information system (GIS). All children diagnosed with type 1 diabetes up to 16 years of age at diagnosis between 1977-1995 were included. The population at risk was obtained directly from the population registry for the respective years and geographical area levels. SETTING: South east region of Sweden containing 5 counties, 49 municipalities, and 525 parishes. MAIN RESULTS: A significant geographical variation in incidence rate were found between the municipalities (p<0.001) but not between the counties. The variation became somewhat weaker when excluding the six largest municipalities (p<0.02). In municipalities with increased risk (>35.1/100 000) the major contribution comes from children in age group 6-10 years of age at diagnosis. There were no obvious differences between the age groups in municipalities with decreased risk (<20.1/100 000). Boys and girls had about the same degree of geographical variation. CONCLUSIONS: Apart from chance, the most probable explanation for the geographical variation in the risk for children and adolescents to develop type 1 diabetes between the municipalities in the region is that local environmental factors play a part in the process leading to the disease.

Islet cell antibodies (ICA), insulin autoantibodies (IAA), islet cell surface antibodies (ICSA) and C-peptide in 1031 school children in a population with a high background incidence of IDDM.

Islet cell antibodies (ICA), insulin autoantibodies (IAA) and islet cell surface antibodies (ICSA) together with C-peptide were determined in 1031 healthy schoolchildren to evaluate the frequency of autoimmune reactions towards endocrine pancreas and its relation to insulin secretion in non-diabetic children. The prevalence of ICA (levels > 6 JDF units) was 1.4% (14/1012) while 44 children (4.3%) were ICSA-positive and 40 (4%) had IAA. Girls had higher titres of ICSA than boys. Young children (7-8 years) more often had IAA than 12-13-year-old children who, however, had ICA three times more often than the young children. There were no clear associations between the different antibodies. Of the children, 2.4% had very low post-prandial serum C-peptide values (< or = 0.25 nmol/l). Serum C-peptide was higher in girls than in boys (P < 0.001) and in older children than in younger (P < 0.001). Girls with low levels of ICA had high C-peptide values, while girls with high ICA titers had low C-peptide values, the latter perhaps indicating partial beta cell loss. IAA and ICSA were not related to C-peptide values but both positive ICSA and high C-peptide values were most common in the autumn (P < 0.02 and P < 0.0001, respectively). One of the ICA-positive children developed diabetes in 1991, 4 years after the blood sample was taken. Since after 5 years only one of the children has developed IDDM, it can be concluded autoimmune reactions towards endocrine pancreas and insulin may occur in many children without the development of manifest diabetes. Those with high ICA titers may have lost so many beta cells that their insulin secretion is affected, which in some cases might lead to diabetes many years later.

Breast-feeding seems to play a marginal role in the prevention of insulin-dependent diabetes mellitus.

In order to test the hypothesis that breast-feeding has a protective effect on the development of insulin-dependent diabetes (IDDM) during childhood we retrospectively studied 297 diabetic children age 15 years or less diagnosed 1974-88 at the 5 pediatric departments in the South-East region of Sweden. They were compared to 792 non-diabetic controls of the same age (year of birth), sex and geographical location. Files from the child welfare clinics were reviewed and the families answered questionnaires. Ten percent of the diabetic children and 11% of the control children had never been breast-fed. In the total material no obvious difference was seen regarding duration of breast-feeding between the diabetic children and their healthy controls. Only in the oldest maternal age group (> 35 years of age) control mothers tended to have a longer duration of breast-feeding (P < 0.01). There was also a trend that children with diagnosis of IDDM during winter had been breast-fed for a shorter period (4.1 +/- 3.4 months) than their controls (5.0 +/- 3.5 months, P < 0.09). We conclude, that breast-feeding has very little effect on preventing IDDM in children. If it has any effect, it might be to decrease the risk in certain subgroups.

A fourfold difference in the incidence of type 1 diabetes between Sweden and Lithuania but similar prevalence of autoimmunity.

We investigated whether other autoimmune disorders in addition to type 1 diabetes are more common in Sweden than Lithuania, and if there are any differences in inheritance patterns of both type 1 diabetes and other autoimmune disorders. Data from 517 children in southeast Sweden and 286 children in Lithuania aged 0-15 years were included in the study. Age- and sex-matched control children were randomly selected. Information was collected by questionnaire. Of the children with diabetes in Sweden, 13.2% had a family member with type 1 diabetes compared to 7% of children with diabetes in Lithuania (P < 0.01) (OR = 2.01). No such difference was seen for other autoimmune diseases in family members of children with diabetes (Sweden 12%, Lithuania 14%, n.s.). Control children in Lithuania had family members with autoimmunity more frequently (15.3%) than control children in Sweden (7.4%, P < 0.001) (OR = 2.26). This difference was most pronounced in mothers. The Lithuanian control children had an autoimmune disease more frequently than the controls in Sweden (4.7% versus 1.5%, respectively, P < 0.001) (OR = 3.21). There seem to be environmental factors that specifically contribute to the development of type 1 diabetes, factors which are less related to the development of autoimmunity in general.

Islet autoantibodies in the prediction of diabetes in school children.

In 1987 serum was collected from 1031 non-diabetic schoolchildren in the Southeast area of Sweden with the aim of evaluating islet autoantibody status (ICA, GADA and IA2-ab) in the prediction of diabetes in schoolchildren. The clinical development of Type 1 diabetes in the children was assessed in 1994 and 1997. The combination of ICA, GADA and IA2-ab were found in four subjects whereas six had two and 35 children one of these antibodies. After 10 years, six of the 1031 children had developed clinical diabetes and five of these six children were positive for islet antibodies. Two were positive for all three antibodies, two were positive for ICA and GADA, and one was positive for GADA. Among the individual autoantibodies, ICA showed the highest positive predictive value (29%) whereas the predictive value for the combination of two autoantibodies was highest for GADA and ICA (40%). Thus, GADA and ICA measurements may be a rational approach to detect schoolchildren at risk for developing diabetes.

Severity at onset of childhood type 1 diabetes in countries with high and low incidence of the condition.

Severity of Type 1 diabetes mellitus (DM) at presentation was compared between south-east Sweden and Lithuania where incidence of childhood Type 1 diabetes is three times lower than in Sweden. New cases of diabetes at age 0-15 years from August 1995 to March 1999 in south-east Sweden and from August 1996 to August 2000 in Lithuania were included. Symptoms and clinical characteristics at diagnosis were recorded. Data about the close environment were collected using questionnaires. Lithuanian children were diagnosed in a more severe condition, mean pH 7.30 and HbA(1c) 11.5% compared with mean pH 7.36 and HbA(1c) 9.7% in Swedish children (P<0.0001). More Lithuanian than Swedish children were diagnosed in ketoacidosis (pH < or = 7.2, hyperglycaemia and ketonuria), 21.3 versus 7.3% (P<0.0001). Only 4.6% of Swedish children and 1.0% of Lithuanian children had no symptoms (P=0.007). Children in families with at least one first degree relative with diabetes (12.2% in Sweden and 8.4% in Lithuania, NS) had laboratory values at diagnosis closer to normal than sporadic cases in either country. Factors predicting ketoacidosis in Sweden were an unemployed mother and absence of infections in the 6 months before diagnosis. In Lithuania it was younger age and mother with less education. Additional educational activities for doctors are needed in countries with low incidence to reduce prevalence of ketoacidosis at onset.