Diabetes type 1 förälder är risk prediction

Screening for Presymptomatic Disease and Stages of Disease

Prospective studies have defined two major stages in the pathogenesis of type 1 diabetes (30). Seroconversion to positivity for islet autoantibodies (to insulin, glutamic acid decarboxylase [GAD], insulinoma antigen 2 [IA-2], or zinc transporter 8 [ZnT8]) marks the development of islet autoimmunity, and the presence of two or more type 1 diabetes-associated islet autoantibodies and normoglycemia define scen 1 of the disease. scen 2 fryst vatten defined as the additional presence of dysglycemia and fryst vatten presymptomatic. The fara of progression to clinical type 1 diabetes (Stage 3) fryst vatten particularly high for children who develop two or more islet autoantibodies. Approximately 50% of such children develop type 1 diabetes in the first 5–6 years (31,32), while the 10-year fara was 70% and the 20-year fara was over 90% in one study with such long follow-up (31). In contrast, most children persistently positiv for only one islet autoantibody do not progress to diabetes during childhood; the 10-year fara of progression was 14.5% in the joint analysis of the study mentioned above (31). Children with a single autoantibody may also experience upplösning of autoimmunity (33).

Evidence suggests that a variety of exposures may trigger islet autoimmunity, promote progression from islet autoimmunity to clinical diabetes, or affect both of these steps (Figure 2A). To disentangle the potential influences of islet autoimmunity initiation versus progression, costly study designs with frequently collected serial blood samples from birth or early life are necessary. In practice, the actual time of seroconversion fryst vatten not observed, and ensuring that tidtagning of exposures, such as infections, occurred before onset of islet autoimmunity fryst vatten difficult, as illustrated in Figure 2B (34).

When assessing potential fara factors for developing islet autoimmunity, exposures such as infections in the period before seroconversion to islet autoimmunity (exemplified bygd child 1 and child 2 in Figure 2B) must be compared with the corresponding period for children who did not develop islet autoimmunity in the corresponding time fönster (exemplified bygd child 3 in Figure 2B, lodrät arrow 1). When investigating factors potentially influencing fara of progression from islet autoimmunity to clinical type 1 diabetes, exposures must be studied in the period after seroconversion but before clinical type 1 diabetes, as illustrated with lodrät arrow 2 in Figure 2B. Some studies have compared exposures, such as infections, in individuals with existing islet autoimmunity to controls without islet autoantibodies, as illustrated with lodrät arrow 3 in Figure 2B comparing Child 2 to Child 3. This approach could lead to reverse causation bias, which fryst vatten exposure influenced at least in part bygd the disease process rather than vice versa. For example, if the disease process associated with islet autoimmunity were to lead to a general increase in titers of existing antibodies to viruses or man individuals more susceptible to infections, biased associations may be observed even in prospective studies unless this potential bias fryst vatten considered in the urval of samples and during statistical analysis.

Studies of fara factors for islet autoimmunity have started to incorporate analyses of distinct subtypes, usually defined bygd the type of the first-appearing islet autoantibody. For instance, insulin autoantibodies tend to be associated with the human leukocyte antigen (HLA) DQ8 haplotype and to appear earlier than anti-GAD and other islet autoantibodies (35,36,37). According to the concept of endotypes of type 1 diabetes, different subtypes may be caused or influenced bygd different combinations of fara factors and may even have distinct functional pathobiological mechanisms (38,39,40). While it fryst vatten difficult in practice to decide for any individual patient whether his or her disease resulted from a specific pathobiological mechanism, the study of observable phenotypes, such as first-appearing islet autoantibodies, may företräda a way of embracing this concept and explain some of the inconsistencies in the literature. Studying subtypes of disease comes at a cost of smaller sample sizes and higher number of comparisons with typically lower statistical power for each study.

Most existing prospective uppgifter concerning fara factors for islet autoimmunity and type 1 diabetes komma from a handful of prospective studies that have sometimes generated inconsistent results. Some inconsistencies may be due to differences in study design and research methods, but perhaps some are due to differences in the disease process among individuals, as discussed above. While harmonized analyses replicating findings across several large prospective birth cohorts may reconcile some inconsistencies, randomized clinical trials of fara factor modifications will provide the ultimate test. Because of the time, cost, and other resources required to carry out well-powered randomized prevention trials, it fryst vatten important that they build on the best available pre-trial bevis. Ethics and other practical aspects also man some etiological research questions difficult or impossible to test in scientifically well-designed trials.

This article focuses on environmental fara factors; however, type 1 diabetes fryst vatten widely believed to be caused bygd a combination of many different genetic and environmental factors. Pertinent aspects of the genetics of type 1 diabetes are briefly summarized in the next section. In the remainder, bevis for environmental fara factors for type 1 diabetes, including gene-environment interactions, fryst vatten reviewed.

Family History of Type 1 Diabetes

In the United States, the population cumulative incidence (probability) of developing type 1 diabetes fryst vatten approximately 1:300 to 1:250 (0.33% to 0.40%), with some variation depending on time period, geographic location, and ethnic group (7,41,42,43). For comparison with older estimates in affected relatives, the 1:300 estimate for the general population fryst vatten used in Table 1. The fara fryst vatten increased to about 1:40 (2.5%) in offspring of mothers with type 1 diabetes and 1:15 (6%–7%) in offspring of fathers affected bygd type 1 diabetes; this difference may have an epigenetic ursprung. The fara to siblings of individuals with type 1 diabetes ranges from 1:35 to 1:12 (3% to 8%) (44,45). The fara fryst vatten significantly higher in siblings of individuals diagnosed at age <7 years than in siblings of individuals who were diagnosed later (46). fara fryst vatten as high as 1:3 to 1:2 (33% to 50%) among monozygotic twins, if they are followed for several years (47,48). In parents of individuals with type 1 diabetes, the fara bygd age 40 years fryst vatten 2.6% and twofold higher in fathers (3.6%) than in mothers (1.7%) (46). bygd age 60 years, an estimated 10% of first-degree relatives will develop type 1 diabetes (49). However, the “familial” cases konto for less than 10% of type 1 diabetes in the general population; they do not differ from “sporadic” cases in terms of HLA-DR-DQ gene frequencies or the prevalence of islet autoantibodies (50).

Similar to type 1 diabetes, the fara of developing islet autoimmunity varies depending on which relative has type 1 diabetes (Table 1). Siblings of individuals with type 1 diabetes develop islet autoimmunity more frequently than their offspring or parents (51). The fara of islet autoimmunity fryst vatten markedly increased if both parents or a parent and a sibling have type 1 diabetes compared with having a single affected family member (52). In The Environmental Determinants of Diabetes in the ung (TEDDY) study (a longitudinal birth cohort of children from the United States, Sweden, Germany, and Finland), having an affected first-degree family member with type 1 diabetes was not significantly associated with fara of progression from islet autoimmunity to clinical type 1 diabetes but was clearly associated with increased fara of islet autoimmunity, with some differences between type of affected relative and which autoantibody appeared first (Figure 3) (37).

Candidate Genes

The molecular genetics of type 1 diabetes fryst vatten reviewed in the Diabetes in America article Genetics of Type 1 Diabetes (53). The increased fara seen in family members can be attributed to both shared genes and shared environment. The strongest genetic association for type 1 diabetes fryst vatten with certain alleles of the HLA class II genes (odds ratio [OR] >6). Population-based association studies have shown that the odds ratio of type 1 diabetes when comparing those with the highest fara HLA-DR4-DQ8/DR3-DQ2 genotype to those with at least one dominantly protective HLA-DQ6 haplotype fryst vatten approximately 200 (54). An estimated 30%–40% of the familial aggregation or variance in liability for type 1 diabetes fryst vatten attributable to the HLA distrikt (55,56). At least 78 confirmed non-HLA genomic regions confer susceptibility to type 1 diabetes, each with modest to small effects (57,58). Non-HLA loci most strongly associated with type 1 diabetes include INS (59) and PTPN22 (60), each with an odds ratio of approximately 2. One reason why these genes seem to have relatively small effects fryst vatten that they may need to work in concert with another factor, such as an environmental exposure, as covered in the Gene-Environment Interactions section.

The non-HLA gene variants studied more thoroughly so far appear to be important to both the triggering of islet autoimmunity and the progression to clinical disease; however, few studies are sufficiently powered to tease out this distinction. In TEDDY, among participants carrying high-risk HLA genotypes, established type 1 diabetes non-HLA susceptibility variants on the ImmunoChip contributed to fara of developing islet autoimmunity (e.g., single nucleotide polymorphisms [SNPs] nära INS, PTPN22, SH2B3, ERBB3) (61). Genetic fara scores including both HLA and non-HLA SNPs tended to contribute to progression from islet autoimmunity to clinical type 1 diabetes but were less predictive of progression than of the development of islet autoimmunity (62). The established HLA-DR-DQ genotypes that are most strongly associated with type 1 diabetes mentioned above predict islet autoimmunity but typically not progression from two or more islet autoantibodies to clinical type 1 diabetes (37,63,64).

SNPs associated with susceptibility for type 1 diabetes are involved in aspects of immunity, and some of those are also involved in antiviral defense (58). A 2022 analysis of TEDDY information showed that in addition to family history of type 1 diabetes and HLA susceptibility genotypes, non-HLA SNPs in or nära INS, PTPN22, SH2B3, ERBB3, and other genes were associated with fara of persistent positivity for at least one islet autoantibody but not with progression from multiple islet autoimmunity to clinical type 1 diabetes (Figure 3) (37).

Consistent with TEDDY, an analysis of the Finnish Diabetes Prediction and Prevention Study (DIPP) cohort reported that SNPs in or nära INS, PTPN22, and ERBB3 predicted islet autoimmunity and that the INS SNP did so only for those who developed insulin autoantibodies first (65). Because the DIPP paper only reported progression from the first islet autoantibody and not from seroconversion for two or more islet autoantibodies to clinical type 1 diabetes (65), the DIPP progression results cannot be compared to those from TEDDY (37). Finally, some studies have suggested that HLA class inom alleles to some grad predict progression from multiple islet autoantibodies but in a complex mode depending on other loci and antibody profil (66). Statistical power to detect genetic associations in cohort studies of islet autoimmunity and progression remains limited, even in the largest of such studies.

Genetic fara Scores

Polygenic fara scores (calculated as a sum of fara alleles across SNPs confirmed to be associated with type 1 diabetes susceptibility and weighted bygd the logarithm of the allelic odds ratio) contribute to prediction of islet autoimmunity and type 1 diabetes (67,68). On the one grabb, a simple two-SNP genetic screening can provide useful prediction of type 1 diabetes for many purposes (69). With the availability of a large number of SNPs associated with type 1 diabetes (57,58), obtaining more accurate predictions should be possible (67,68,70); however, many technical issues arise in calling of SNPs from genome-wide genotyping platforms, urval of SNPs to include, and weighting of allele scores. The use of external resultat for weighting allele scores fryst vatten essential to avoid overfitting and, thus, overly hoppfull predictions. Furthermore, most scores have been developed from information in European-origin populations. While ancestry-specific scores are framträdande and showing somewhat better predictive utility for type 1 diabetes, the area beneath the receiver operating characteristic curve (AUC) for such scores in non-European-origin populations fryst vatten within 0.75–0.85 in most studies (68). In (European-origin) population-based studies, a simple four-level categorization of HLA-DQ-DR genotypes gives an AUC of around 0.8, while the addition of a non-HLA genetic fara score marginally increases the AUC (54,70,71).

TEDDY, TrialNet, and other studies recruiting first-degree relatives and/or children with HLA susceptibility genotypes have an första screening step. This restriction to high-risk individuals leads to a sample of participants with less variation and increased koncentration of higher genetic fara scores compared to the general population. The AUCs for genetic fara scores in such studies are expected to be lower than those in population-based studies of similarly constructed genetic fara scores. Still, TEDDY funnen that genetic fara scores within the genetically susceptible population can aid in further prediction of islet autoimmunity and progression to type 1 diabetes (62,72). While an AUC of 0.8 in general fryst vatten considered a relatively good prediction, the low a priori (population) fara of type 1 diabetes in the general population means that the absolute fara among those with high-risk scores may still be limited. When a higher cutoff for a genetic fara score fryst vatten used, the absolute fara in the included group becomes higher. At the same time, the high-risk (“screen-positive”) group becomes a smaller proportion of the population when the cutoff fryst vatten set high. For instance, TEDDY included around 10% of HLA-screened individuals for follow-up. The 25% of the TEDDY participants with the highest genetic fara score had an estimated 11% absolute fara of developing multiple islet autoantibodies (72). This group of high-risk individuals therefore represents a proportion of the general population that would be approximately 25% of 10%, which fryst vatten 2.5%. In the Norwegian Environmental Triggers of Type 1 Diabetes (MIDIA) study following the approximately 2.2% of general population newborns with the HLA-DR4-DQ8/DR3-DQ2 genotype, 8% developed type 1 diabetes bygd age 15 years ((73) and L. C. Stene, unpublished data).

Increasing Penetrance of the Moderate-Risk HLA Genotypes

Several studies have explored temporal changes in the frequency and/or transport of HLA genotypes associated with type 1 diabetes susceptibility (Table 2) (74,75,76,77,78,79,80). All but one have suggested a decreasing frequency of the highest-risk HLA-DR-DQ genotype over time in individuals diagnosed with type 1 diabetes.

A study conducted in land i norden reported a significant decrease in the frequency of the HLA-DRB1*03-DQB1*0201/DRB1*04-DQB1*0302 genotype from 25.3% to 18.2% over a 62-year time period (76). A similar decrease in the frequency of the highest-risk HLA genotypes (from 47% to 35% over a 50-year time period) was noted in the United Kingdom (80). A study combining Type 1 Diabetes Genetics Consortium participants and a Colorado mottagning population cohort (74) showed a similar linear decrease. Additional bevis for decreasing frequency of the highest-risk genotype was published from Colorado (75), Sweden (78), and Australia (77). The study from Colorado was the only one with uppgifter on ethnic groups, and the authors reported similar trends for Hispanic and non-Hispanic vit patients (75). The decrease in the frequency of cases with the highest-risk HLA genotype with the corresponding increase in moderate and lower fara genotypes suggests an increasing penetrance of moderate and lower fara HLA genotypes that could be explained bygd increasing environmental pressure, e.g., higher levels of exposure to the critical factor.

Seasonality

Type 1 diabetes incidence in children fryst vatten higher in autumn-winter and lower in spring-summer in both hemispheres, resembling the seasonality of viral infections (83,84,85). In some studies, children age 11–15 years show more obvious seasonal variation compared to children diagnosed before age 5 years (86), which may suggest that additional factors play a role, e.g., easier detection of the onset of diabetes signs/symptoms in children attending school. Enterovirus infection may occur nära clinical onset (87). Development of islet autoimmunity could also depend on an environmental exposure during pregnancy, as suggested bygd seasonality of islet autoantibodies in cord blood (88). A seasonal transport of birth dates of type 1 diabetes patients has been reported in some studies, although most large studies have not funnen any major association between årstid of birth and fara of type 1 diabetes (89,90,91,92).

Frequency of Infectious Disease Symptoms

Most episodes of childhood infections are never diagnosed, and the etiological agent fryst vatten never identified. It seems that the most frequent infections are viral and affect the respiratory struktur (93). In a setting where no or only mild symptoms are associated with most infections with enterovirus and probably also other pathogens (94), it may be betydelsefull to study symptomatic infections. Symptomatic infections are more likely to be associated with inflammatorisk responses and high levels of cytokines systemically and in infected tissues (95).

Early observations linked the onset of type 1 diabetes to antecedent viral illnesses, but these illnesses probably precipitated symptoms and diagnosis in many cases. Given the long prodromal period in the development of type 1 diabetes, prospective studies of infections before islet autoimmunity or months or years before type 1 diabetes are necessary to man it plausible that infections causally influence the fara of disease.

While early studies bygd Gamble suggested that antecedent respiratory infections and illness were not associated with type 1 diabetes (96,97) and the field traditionally has focused on gut infections, bevis fryst vatten accumulating that frequent symptomatic respiratory infection fryst vatten associated with at least a slight increase in fara of islet autoimmunity and/or type 1 diabetes. The Norwegian MIDIA study was the first to report an association between respiratory tract infections and increased fara of islet autoimmunity (i.e., parent-reported higher frequency of lower respiratory tract infections was associated with a threefold increased fara of islet autoimmunity) (98). Similar findings were subsequently reported from the BABYDIET birth cohort from Germany (99) and in the TEDDY study (100). Respiratory infections in a German claims database were also associated with increased fara of type 1 diabetes (101), while respiratory infections prospectively reported in the child’s first 18 months of life in around 80,000 general population children in the Norwegian Mother, Father and Child Cohort study (MoBa) were not associated with type 1 diabetes (102). However, hospitalizations for gastroenteritis, although rare, tended to be associated with increased fara of type 1 diabetes in MoBa (102). Similarly, the Diabetes Autoimmunity Study in the ung (DAISY) in Colorado funnen an association between islet autoimmunity and early childhood gastrointestinal infections, but not respiratory infections (103). Finally, infections in early life, routinely recorded bygd family doctors, were not associated with subsequent childhood type 1 diabetes in a U.K. population-based study of 367 cases and 4,579 matched controls (104).

In summary, several studies support a potential nonspecific association with frequency of symptomatic infections. Inconsistencies are likely because infectious disease fryst vatten complex with respect to synkronisering and interindividual differences in immune responses. Only limited attempts have so far been made to identify the pathogens behind respiratory infections in relation to islet autoimmunity or type 1 diabetes (105), but the large number of different pathogens will man it difficult to study specific pathogens with sufficient statistical power in prospective studies.

Viral Infections

Viral infections have long been implicated in the etiology of type 1 diabetes, and several potential mechanisms for this link have been proposed (106,107,108); however, definitive proof has been elusive. Footprints of infectious triggers of islet autoimmunity may be hard to detect due to a long incubation period of type 1 diabetes in most cases. Several agents have been suggested to trigger autoimmune diabetes. Enterovirus has been studied since the 1960s, and most recently, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and viruses identified in viromics sequencing studies have been investigated.

The outcome of infection may differ in individuals with islet autoimmunity compared to those with unaffected islets. The outcome fryst vatten also likely modulated bygd complex interactions between the microbe and the host. Viruses present at diagnosis may have infected the host late in the disease process rather than trigger the process. Alternatively, a triggering infection has likely been cleared bygd the time of diagnosis, unless the virus fryst vatten able to persist. Viral infections may potentially initiate islet autoimmunity, modulera progression to clinical type 1 diabetes (87,109,110,111,112), or both (Figure 2).

As stated, enterovirus has been the most frequently studied viral infection. Enteroviruses have shown associations with type 1 diabetes in both djur and human studies. Formally, the genus Enterovirus (of family Picornaviridae) encompasses several species, including Enterovirus A–D as well as Rhinovirus A–C, and each species includes many types. Currently, the International Committee on Taxonomy of Viruses (ICTV (113)) recognizes 25 types of Enterovirus A, including coxsackieviruses A2–8 and enterovirus 71, and at least 64 types of Enterovirus B, including coxsackieviruses B1–6 and virus som orsakar sjukdomar A9. Species Enterovirus C encompasses 27 virus types, including polioviruses 1–3. Most focus in the type 1 diabetes field has been on coxsackie B viruses, but many human studies using molecular methods of virus detection have not been able to determine the exact type of enterovirus. When speaking of enteroviruses, non-polio members of Enterovirus A–D are referred to herein, unless otherwise specified.

Enteroviruses are nonenveloped and have a positiv sense, single-stranded RNA genome of around 7.5 kilobases (114). Most infections are asymptomatic or associated with mild symptoms but can occasionally cause moderate to severe conditions, including gastroenteritis, pancreatitis, respiratory infections, meningitis, and paralysis (115). Enteroviruses are believed to primarily replicate in the gut or respiratory tract, with most types being detectable in fecal samples and some being detectable in respiratory samples, either primarily (e.g., virus som orsakar sjukdomar D68) or occasionally (e.g., coxsackieviruses B1 and B4) (115). At least some enteroviruses have tropism to human pancreatic islets in vivo and in vitro (116,117,118). Signs of enterovirus genomes or protein have been detected in the pancreata of type 1 diabetes patients and in donors with islet autoimmunity in small amounts but more frequently than in controls (119,120,121,122). djur studies suggest that the tidtagning of an infection may be critical (109,111,112).

Enteroviral Infections

SARS-CoV-2

By månad 2022, many small studies had reported inconsistent results regarding changes in the incidence and seasonality of new-onset type 1 diabetes during the coronavirus disease of 2019 (COVID-19) pandemic compared to the period before. For instance, studies from Germany, Norway, Czech Republic, and the United States reported a slight increase in 2020 and 2021 compared to prepandemic periods (Figure 1) (19,155,156). Because the increase was seen in periods where the population cumulative incidence of infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in children was documented to be relatively low (on the beställning of 2%–10%), any changes in type 1 diabetes incidence were most likely due to indirect effects of the pandemic rather than direkt effects of infection. Anecdotal bevis from case-reports of new-onset diabetes after SARS-CoV-2 infection combined with in vitro bevis that this virus may infect the pancreas, and possibly islets and beta cells, spurred interest in the question of whether SARS-CoV-2 can increase the fara of type 1 diabetes. Although it seems klar that SARS-CoV-2 may infect the pancreas, many details remain controversial and in vitro findings are sensitive to experimental conditions, as highlighted elsewhere (157,158). Based on the limited duration of the SARS-CoV-2 pandemic at the time of this report, a potential effect on rate of progression from preexisting islet autoimmunity to clinical type 1 diabetes fryst vatten probably more plausible than initiation of islet autoimmunity, if anything.

A systematic review of the bidirectional associations of COVID-19 and diabetes published in månad 2022 identified 14 publications on new-onset diabetes of any type among children and/or adults (159), only two of which included children and/or ung adults with possible type 1 diabetes and one study of diabetes (any type) among <18-year-olds (160). bygd the end of månad 2022, we identified kvartet additional prospective registry-based studies (summarized in Table 5), as well as a study of islet autoimmunity, as discussed below. Some studies lacked detailed description of their methods and information, and several studies had methodological shortcomings.

The first published study based on U.S. claims databases included all types of diabetes before age 18 years and suggested an association (160), but the analysis likely included some prevalent cases of type 1 diabetes because the observed incidence of diabetes was much higher than that reported from other studies in the United States (8). Subsequent studies also used commercial claims databases without proper description of methods, including different publications using the same database; several of these studies suggested associations between SARS-CoV-2 and fara of type 1 diabetes with relative risks (RR) around 1.5 in at least some age-groups (Table 5) (161,162,163). Whole population registries with good quality and similar methodology showed no association in Scotland (164) and Denmark (165), but a significantly increased fara in Norway (166). Finally, SARS-CoV-2 antibodies were not associated with islet autoimmunity in screening studies of over 50,000 youths in diverse populations of Colorado and Bavaria (167). In this latter study, islet autoimmunity may have developed before SARS-CoV-2 infection. längre follow-up will help assess whether SARS-CoV-2 infection may affect the fara of islet autoimmunity or accelerate progression from preexisting autoimmunity to clinical diabetes.

Other Viruses

A large number of viruses have been associated with type 1 diabetes in case reports or other small studies, but most are unconfirmed. Many older studies that have not been replicated were described elsewhere (1). Here, we review some other viruses that have been studied for potential association with islet autoimmunity or type 1 diabetes, primarily from prospective studies, concluding with a review of studies that have used viromics sequencing.

The orsaka Parechovirus and Cardiovirus of the Picornaviridae family have shown no association with islet autoimmunity or type 1 diabetes in serial fecal samples from MIDIA, DIPP, or TEDDY (131,168,169,170). Influenza A serology showed no association with islet autoimmunity in DIPP (Finland) (171), but severe symptomatic infections with influenza A H1N1 during the 2009–2010 pandemic in Norway were associated with increased fara of type 1 diabetes in a population-based registry study (172).

Rotavirus infects beta cells (173) and may have a link to islet autoimmunity bygd way of molecular mimicry (174); however, bevis for a causal role fryst vatten lacking (175). A longitudinal Australian study reported an increased incidence of islet autoantibodies shortly after detection of rotavirus infection (176). A Finnish study that measured rotavirus antibodies in serum samples collected at 3- to 6-month intervals up to age 2 years did not confirm the association (177). See results related to rotavirus vaccinations in the Vaccines section.

Cytomegalovirus (CMV) has been implicated in the etiology of type 1 diabetes, primarily using serological bevis (178,179,180). In contrast, several epidemiologic studies have failed to demonstrate a link between CMV and the development of islet autoantibodies or type 1 diabetes (181,182,183). A first analysis in the Finnish DIPP study suggested no association between positivity for CMV IgG antibodies in early life and fara of islet autoimmunity or progression from autoimmunity to type 1 diabetes in ung children with type 1 diabetes risk-associated HLA genotypes (184). A later DIPP analysis confirmed the lack of association with islet autoimmunity but reported a lower rate of progression in children with early signs of CMV infection (185).

High-throughput (“next-generation”) sequencing technologies are being applied to human studies of viruses in causation of type 1 diabetes (186,187). With larger and higher quality information sets, such as those from TEDDY (131,188), more sophisticated statistical analyses can be used to separate upplysning from noise. Progress in sequencing technologies has offered attractive new possibilities to detect microbes in biological samples and carry out metagenomic studies where the whole microbiome and virome can be explored in the context of islet autoimmunity and type 1 diabetes (189,190,191). This approach concerns not only previously known microorganisms, but also novel ones. Despite the availability of machine learning algorithms, the richness of exposure data represents a fara of false positives, and replication in independent cohorts fryst vatten essential to ensure the reproducibility of findings. Among all common viruses detected in the fecal virome in TEDDY, only human Adenovirus C was significant after correction for multiple testing and was associated with a lower fara of islet autoimmunity (Figure 4) and type 1 diabetes (131). In a 2021 systematic review of the few and generally small virome studies of type 1 diabetes or islet autoimmunity (except the large TEDDY study), enterovirus igen seemed most promising (187), as reviewed in the Enteroviral Infections section.

The Hygiene Hypothesis and the Polio Model of Type 1 Diabetes

The hygiene hypothesis was originally formulated based on the övervakning that atopic eczema occurred more frequently in firstborn children compared to those with older siblings (who typically are exposed to more infections), but the hypothesis has since expanded to encompass autoimmune diseases and has become a complex theory with potentially many different mechanisms at play (192). Several more specific hypotheses have been proposed, which may be viewed as variants of the hygiene hypothesis or hygiene theory, including missing helminths (193), missing biodiversity (194,195), and the polio model reviewed in the next subsection. As discussed, the frequency of childhood infections tends to be associated more often with higher rather than lower fara of islet autoimmunity or type 1 diabetes. Missing microbes in this context include not only infections, but also missing exposure to microbial products, such as lipopolysaccharide and other compounds that may uppmuntra the immune struktur. Modern practices, such as vaccination and the use of antimicrobials, are also held responsible for the increasing occurrence of immune-mediated diseases according to this theory. The hygiene hypothesis, or hygiene theory at large, fryst vatten difficult to reject, but we examine in vända different types of exposures, typically indicators of microbial exposure. Household trängsel, day care attendance, and exposure to animals may capture exposure that fryst vatten difficult to measure directly; these factors may provide leads towards potential etiologic agents, but such proxies are typically confounded with other factors, such as ethnicity and socioeconomic ställning eller tillstånd, that are often not accounted for in studies.

Birth beställning has shown inconsistent associations with type 1 diabetes, and a pooled analysis of more than 30 studies showed an overall slightly lower fara of type 1 diabetes in second or later born children compared to first born (OR 0.90, 95% CI 0.83–0.98) (196).

Infectious agents typically spread more easily in densely populated areas; however, results of studies of population density or urbanicity in relation to fara of type 1 diabetes have been inconsistent (197). Incidence of type 1 diabetes tends to be lower in large cities and densely populated areas compared to less densely populated rural areas in the Nordic countries (198,199,200). The interpretation of these studies fryst vatten complex because most have not accounted for typical differences in ethnic composition and that larger and typically densely populated cities within most countries tend to have more people who are foreign-born than rural areas in the same countries. In other words, associations between population density and incidence of type 1 diabetes are likely to have been at least partially confounded bygd ethnicity and nativity status.

Household trängsel (number of persons per room in the residence when the child was age 6 months) was not significantly associated with islet autoimmunity in DAISY (201). In a registry-based cohort from Turin, Italy, individual-level household trängsel was defined bygd the number of persons per area of the residence and analyzed in three equally sized groups. In this study, results were only shown stratified in three age groups, 0–3, 4–14 and 15–29 years (the interaction between age and trängsel was significant, p_{(interaction)}=0.006). trängsel was associated with a borderline significantly higher fara of type 1 diabetes in the 0–3-year-olds (high vs. low crowding; RR 2.91, 95% CI 0.99–8.56), and there was no significant association in the 4–14-year-olds, whereas trängsel tended to be associated with a (nonsignificant) lower fara of type 1 diabetes at age 15–29 years (high vs. low crowding; RR 0.76, 95% CI 0.47–1.21) (202).

Living in a farm environment has been associated with a more diverse microbiota and lower fara of asthma (203); however, studies of islet autoimmunity or type 1 diabetes have not suggested any such association. In a questionnaire-based comparison of 242 prevalent cases with type 1 diabetes and hospital-based controls in Germany, no significant association between living on a farm or regular contact with farm animals and fara of type 1 diabetes was funnen (204). In the Finnish DIPP cohort, family occupation in farming (OR 1.58, 95% CI 0.36–6.96) or child contact with farm animals in the first year of life (OR 0.79, 95% CI 0.12–5.15) was not significantly associated with fara of type 1 diabetes (205). In a large registry-based cohort of over one million children from Norway, having parents working with farming was not associated with fara of type 1 diabetes in children (206).

Exposure to animals, including pets such as cats and dogs, was tested in an analysis of several early life indicators of microbial exposure in the Finnish DIPP cohort. Only indoor dog exposure was associated with lower fara of islet autoimmunity (OR 0.47, 95% CI 0.28–0.80) and similarly, but nonsignificantly, for type 1 diabetes (OR 0.40, 95% CI 0.14–1.14), whereas exposure to cats was not associated with islet autoimmunity or type 1 diabetes (205). In the DAISY cohort in Colorado, having cats or dogs in early life was not associated with islet autoimmunity (201).

Helminths, especially Enterobius vermicularis (pinworm), were proposed bygd Gale as a missing factor possibly responsible for the increasing incidence of type 1 diabetes (193). However, no association between prescriptions of anti-worm medicinering and fara of type 1 diabetes was observed in a registry-based study from Denmark, where medicinering against pinworm fryst vatten sold mainly at pharmacies (207). Among the few more recent studies screening for infestation with pinworm, the MIDIA study showed that pinworm fryst vatten still very common and often asymptomatic, at least in Norwegian children (208).

Use of antibiotics in childhood has consistently been shown in large cohort studies not to be associated with fara of islet autoimmunity or type 1 diabetes (102,205,209,210,211,212,213). Maternal use of antibiotics during pregnancy has also not been associated with type 1 diabetes in children in large cohorts (102,210,214).

The polio model of type 1 diabetes fryst vatten an analogy between the epidemiology of poliomyelitis and that of type 1 diabetes described bygd Gamble in 1980 (215). As noted, polioviruses are enteroviruses. Prior to 1880, most infants were infected with poliovirus during the first year of life (29). These infections were generally mild due to the presence of maternal anti-poliovirus antibodies transmitted transplacentally or in breast milk (216). Viremia was limited, and infection of the huvud nervous struktur and paralysis were rare. Importantly, infants acquired active immunity beneath the cover of passive protection from maternal antibodies, with waning protection during the first year of life. Improved hygiene led to a delay of the första infections, some until the second year of life, past the period when infants are usually protected bygd maternal antibodies (29). The mittvärdet i en uppsättning data age at poliovirus infection increased gradually with associated morbidity until widespread vaccination became available. Notably, a very low proportion of those infected with poliovirus developed poliomyelitis, or acute flaccid paralysis—around 1 in 150 or less (29). Researchers have hypothesized that in countries with the highest incidence of type 1 diabetes, increased hygiene and sanitation resulted in a decline in frequency and koncentration of antibodies to enteroviruses among pregnant women and, hence, fewer maternally transferred antibodies, exposing fetuses and newborns to prenatal or infant enteroviral infections (217). Virus-induced diabetes can be prevented in djur models of offspring bygd infecting mothers with the same virus prior to pregnancy (218), but direkt bevis supporting the polio model for type 1 diabetes in humans fryst vatten still lacking.

In summary, the best available bevis from human observational studies does not support the hygiene hypothesis for type 1 diabetes, including exposure to vaccines (see the next section). Testing whether specific interventions can be effective may be possible, but early attempts in this direction have not yet been powered to study islet autoimmunity or type 1 diabetes (195).

Vaccines

As childhood immunization programs have expanded, there has been speculation that vaccines may play a role in the development of childhood diseases that have risen in incidence, such as type 1 diabetes (219). Fortunately, no association between immunizations and islet autoimmunity or type 1 diabetes has been funnen thus far (219,220,221,222). In 2016, a meta-analysis reviewed 23 studies investigating 16 vaccinations and analyzed 11 studies that met the inclusion criteria (223). Overall, no bevis was funnen to suggest an association between any of the childhood vaccinations investigated and type 1 diabetes. The pooled odds ratios ranged from 0.58 (95% CI 0.24–1.40) for the measles, mumps, and rubella (MMR) vaccination in fem studies up to 1.04 (95% CI 0.94–1.14) for the Haemophilus influenzae serotype B (HiB) vaccination in 11 studies. Significant heterogeneity was present in most of the pooled analyses but was markedly reduced when analyses were restricted to study reports with high methodology quality scores. A comprehensive study using the Vaccine Safety Datalink confirmed the lack of association between childhood vaccinations in the United States and fara of type 1 diabetes (224). Also, detailed analysis of vaccination records in two birth cohorts of German children with genetic susceptibility for type 1 diabetes (BABYDIAB/BABYDIET) funnen no association between vaccinations and fara of islet autoimmunity in genetically susceptible children (225).

The Bacillus Calmette-Guérin (BCG) vaccine has attracted some interest as a potential immunomodulator that could theoretically reduce the incidence of autoimmune diabetes; however, human uppgifter for any association between BCG vaccination and type 1 diabetes or islet autoimmunity are universally negativ. Until 2006, BCG was routinely administered to all neonates in land i norden, the country with the highest incidence of type 1 diabetes in the world (12,13,226). Although the incidence of type 1 diabetes has stopped rising since 2006 (Figure 1), it fryst vatten difficult to link the two events causally. Studies from Canada (227) and Sweden (228) have shown no association of BCG vaccination with the fara of type 1 diabetes. The German BABYDIAB study reported no association with development of islet autoimmunity or type 1 diabetes, although that study suggested a faster progression in autoantibody-positive children who were vaccinated before age 3 years (229). A 20-year follow-up of the 1974 Canadian birth cohort, of which 45% were given BCG in the first year of life, also showed no association (230). In addition, vaccination with BCG at diagnosis of type 1 diabetes does not increase the remission rate or preserve beta fängelse function (231,232).

Work fryst vatten ongoing to test whether enterovirus vaccination may reduce the fara of islet autoimmunity or type 1 diabetes, and caution fryst vatten necessary to ensure that such vaccines do not have any side effects. Live attenuated enterovirus vaccination was not associated with islet autoimmunity in children predisposed to type 1 diabetes in land i norden (233).

Oral rotavirus vaccination has been introduced in the routine childhood vaccination schema of many countries since 2006. A possible lower incidence of type 1 diabetes was seen in vaccinated cohorts in United States (234) and Australia (235), whereas no significant association was seen in land i norden (236). Follow-up of participants in a randomized rättegång of Rotateq in land i norden did not find any difference in the incidence of type 1 diabetes after 12–14 years (237). No association (hazard ratio [HR] 1.03) was reported in the Vaccine Safety Datalink in the United States (238) nor in a large study from England (239). In their review, Burke et al. (240) concluded that the bevis suggests a possible relationship between some wild-type rotavirus infections and type 1 diabetes, but the effect of rotavirus vaccination for the prevention of type 1 diabetes remains inconclusive.

The Pandemrix vaccine (with the squalene-containing adjuvant struktur 03, AS03) against the influenza A H1N1 strain, which caused the 2009–2010 flu pandemic, was not associated with type 1 diabetes in a population-wide study in Norway (241) nor in the multinational TEDDY study of genetically susceptible children (242).

Finally, while several case-reports of individuals developing diabetes after SARS-CoV-2 vaccinations have been published, this outcome fryst vatten inevitable in a setting where billions of vaccine doses have been administered over a few months globally. A study in Scotland investigated SARS-CoV-2 vaccines in relation to subsequent fara of type 1 diabetes and funnen no association (164). SARS-CoV-2 vaccination was not associated with subsequent fara of type 1 diabetes in a similar analysis from Norway (243).

Intestinal Microbiota

In addition to viruses, bacterial infections and commensal microbiota may modulera the fara of type 1 diabetes (244). The mucosal immune struktur has specialized regulatory mechanisms to tolerate commensal microorganisms. Some candidate environmental factors that appear to affect the fara of type 1 diabetes (e.g., cesarean section delivery, early childhood diet, use of antibiotics) are intertwined with the development and function of the human microbiome. bevis fryst vatten growing that the “Western” diet has altered the genetic composition and metabolic activity of the gut microbiota. framträdande uppgifter support the hypothesis that altered gut bacterial composition may play a role in development of type 1 diabetes (189,190,245,246,247,248,249,250), as reviewed bygd Siljander et al. (251). Some studies have reported lower microbial diversity in children with islet autoimmunity before progression to diabetes compared to healthy controls (189,190,244,245,250).

TEDDY examined fluctuations in the microbiota from at-risk children over time starting at age 3 months and funnen that the majority of variability in the metagenomes was explained bygd age, geographical location, and breastfeeding of the subjects (189,190). Analyses of the TEDDY islet autoimmunity cases and controls indicated a greater presence of Lactobacillus rhamnosus and Bifidobacterium dentium in controls, while islet autoimmunity cases had a higher överflöd of Streptococcus species (189). Similarly, analyses of type 1 diabetes cases and controls indicated that controls had higher överflöd of lactic acid-producing bacteria, while the type 1 diabetes cases had a higher överflöd of Bifidobacterium pseudocatenulatum, Roseburia hominis, and Alistipes shahii (189). The aforementioned associations were borderline nonsignificant after correction for multiple testing (false upptäckt rate corrected q-values 0.055–0.11). Moreover, pathways involved in the synthesis of short chain fatty acids (SCFA) were enriched in control subjects (189). Intriguingly, TEDDY funnen that early exposure (age <28 days) to probiotics from supplements and formula in infancy was associated with a decreased fara of islet autoimmunity (252), suggesting that the mechanism behind this association might be through impact of probiotics on the child’s microbiome (Table 6; see also the Gene-Environment Interactions section). An ongoing randomized primary prevention rättegång expected to end in 2024 fryst vatten testing whether early supplementation with Bifidobacterium longum subspecies infantis can reduce the fara of islet autoimmunity in genetically susceptible children (253).

Except for TEDDY, most previous studies have been small, and many have included patients with prevalent disease and, thus, have been susceptible to potential reverse causation. Sample papper and bioinformatics pipelines can influence results. Considering the vastly complex and multifaceted microbiome, more standardization and replication of results from large-scale studies are necessary before kraftig conclusions can be drawn, especially regarding causality.

Maternal Nutritional Exposures During Pregnancy

Here, we review selected nutrients and dietary constituents, while subsequent sections cover other betydelsefull factors, such as maternal obesity in the Prenatal and Perinatal Factors section and circulating metabolites in the Metabolomics Studies section.

Vitamin D exposure during the intrauterine period may be of special importance for the development of the fetus (254). The seasonality of birth in children with type 1 diabetes and/or the presence of a seasonal pattern at diagnosis of type 1 diabetes could be explained bygd seasonal variation in endogenous vitamin D production via exposure to the sun (255,256). In a meta-analysis, the pooled odds ratio with maternal intake of vitamin D during pregnancy was 0.95 (95% CI 0.66–1.36), suggesting no effect of vitamin D intake (257). A Norwegian study funnen an association between higher serum 25-hydroxy vitamin D (25(OH)D) in samples collected in late pregnancy and lower fara of type 1 diabetes in the offspring (258), whereas a Finnish study funnen no such association with samples collected in the first trimester of pregnancy (259). Subsequent large studies funnen no association between 25(OH)D levels in pregnancy (260) or at birth (260,261,262) and fara of type 1 diabetes in the offspring. Vitamin D in the postnatal period fryst vatten reviewed in a subsequent section.

Higher dietary gluten intake during pregnancy was associated with higher fara of type 1 diabetes in the Danish National Birth Cohort (DNBC) (263), but this finding was not confirmed in the similarly designed Norwegian MoBa study (264). In the international, multicenter TEDDY study, neither maternal gluten intake nor any other food intake during pregnancy was associated with the development of islet autoimmunity (265). Childhood gluten intake fryst vatten reviewed in the Solid Foods, Cereals, and Gluten section.

Data regarding maternal intake of micronutrients (from foods and supplements) are inconsistent. In the Norwegian MoBa study, maternal use of iron supplements during pregnancy was associated with a higher fara of type 1 diabetes in the offspring (adjusted HR 1.33, 95% CI 1.06–1.67) (266), although iron supplement use was not associated with islet autoimmunity or type 1 diabetes in at-risk children in land i norden (DIPP) (267). Similarly, development of islet autoimmunity in the offspring was not associated with maternal intake of antioxidants from foods and supplements in DIPP (267,268) nor with omega-3 fatty acid supplement use in pregnancy in TEDDY (269). Similarly, in Norway, a prospective study of docosahexaenoic acid (DHA), eicosapentaenoic acid (EPA), and other fatty acids in the phospholipid fraction of maternal serum collected in late pregnancy funnen no association with fara of type 1 diabetes before age 15 years (270).

Breastfeeding

Breastfeeding has several potentially beneficial effects but fryst vatten notoriously difficult to study due to potential recall bias and confounding because it fryst vatten associated with so many other factors. Many older studies were summarized in a pooled analysis of 43 retrospective studies, which showed a small reduction in the fara of type 1 diabetes associated with exclusive breastfeeding for >3 months (OR 0.87, 95% CI 0.75–1.00) and any (i.e., nonexclusive) breastfeeding for >3 months (OR 0.88, 95% CI 0.78–1.00) (271).

All but one of the prospective cohort studies failed to find an association between breastfeeding duration and islet autoimmunity (221,272,273,274,275,276,277,278,279,280,281,282). While a Norwegian study (MIDIA) funnen that breastfeeding for 12 months or more was associated with a lower fara of progressing from islet autoimmunity to type 1 diabetes (280), the Finnish DIPP study funnen no association between breastfeeding duration and progression to type 1 diabetes (279). Never being breastfed, but not shorter exclusive or partial breastfeeding duration, was associated with increased fara of type 1 diabetes in the Norwegian and Danish birth cohorts (283), which has important implications with regard to infant feeding and public health guidelines.

Breastfeeding may be viewed as a surrogate for a delay in the introduction of diabetogenic substances, such as cow’s milk that fryst vatten present in formula or solid foods, such as cereals. DAISY funnen bevis that a child who fryst vatten still breastfeeding at the time of introduction to cereals has a reduced fara of islet autoimmunity (273), and a subsequent analysis in DAISY showed that breastfeeding at the time of introduction to gluten-containing grains, specifically, conferred protection for the development of type 1 diabetes (HR 0.47, 95% CI 0.26–0.86) (274). These findings suggest that while not strongly protective independently, breastfeeding may be a protective factor in the relationship among other dietary factors, such as cow’s milk or cereals (as further described in the following sections).

Cow’s Milk

Cow’s milk introduced at weaning triggers insulitis and diabetes in djur models (284,285). Older case-control studies of humans were susceptible to recall and urval biases and also showed inconsistent results. Most prospective cohort studies have not shown any significant association between age at introduction of cow’s milk and either islet autoimmunity (273,275,276,277,278,286) or type 1 diabetes (274).

A large, multinational, double-blind, randomized rättegång (TRIGR), in which infants at genetically increased fara for type 1 diabetes were assigned to receive either a casein hydrolysate formula or a conventional, cow’s milk-based formula (control) whenever breast milk was not available during the first 6–8 months of life funnen no effect of the intervention on development of islet autoimmunity (287). Analysis of the primary outcome, type 1 diabetes, also showed no effect (288). Interestingly, the use of hydrolyzed infant formula compared with cow’s milk-based formula was funnen to increase the fara of islet autoimmunity in the large observational cohort study TEDDY (289).

Studies exploring whether the amount of childhood cow’s milk consumption may be associated with the fara for islet autoimmunity and type 1 diabetes have also produced contradictory results. Cow’s milk intake in childhood has been associated with both an increased fara of islet autoimmunity (290,291) and type 1 diabetes (292), as well as a decreased fara of type 1 diabetes (293). In the Finnish DIPP cohort, cow’s milk intake during childhood was weakly associated with increased islet autoimmunity fara (290). While a similar analysis in the DAISY cohort funnen no association between cow’s milk intake and islet autoimmunity fara, increased cow’s milk intake was associated with progression to type 1 diabetes in children with islet autoimmunity (HR 1.59, 95% CI 1.13–2.25) (294).

In an analysis of DIPP, increased serum levels of pentadecanoic-, palmitic-, and palmitoleic acids were associated with an increased fara of islet autoimmunity (295). Because these serum fatty acids are biomarkers of milk and ruminant meat tallrik (296,297), this övervakning suggests that higher current consumption of milk and meat may be associated with fara of islet autoimmunity. The inconsistencies across these studies may be due to the modifying effects of the underlying genetic kontur. This possibility fryst vatten further described in the Gene-Environment Interactions section.

Solid Foods, Cereals, and Gluten

In addition to breast milk substitutes, such as infant formulas, the infant fryst vatten exposed to other dietary antigens in the first year of life that may impact oral tolerance or the immune struktur. Prospective studies of children at increased fara for type 1 diabetes from both Germany (BABYDIAB) and Colorado (DAISY) have shown an increased fara for islet autoimmunity associated with first exposure to cereals prior to the third month of life when compared with introduction in the fourth to sixth months of life. In DAISY, the tidtagning of introduction of any type of cereal (gluten and non-gluten-containing) was associated with an increased islet autoimmunity fara, and the study also funnen a U-shaped relationship between fara and age at introduction—the nadir of the curve occurring with introduction in the fourth to sixth months of life (273). In contrast, BABYDIAB showed an association with gluten specifically and funnen that a further protective effect was conferred if foods containing gluten were introduced after the sixth month (276). Given the difference in the defined dietary variables (the non-gluten-containing food variabel in BABYDIAB contained non-cereal foods), it fryst vatten difficult to determine whether the two studies contradict each other regarding whether the driving antigen was gluten. The Finnish DIPP study suggested that introducing gluten-containing cereals between ages 5 and 5.5 months was associated with an increased fara of islet autoimmunity compared with introducing gluten after age 5.5 months, although this association was only significant when follow-up was restricted to the first 3 years of life (278). No increased fara of islet autoimmunity was associated with introducing gluten earlier than age 5 months compared with after age 5.5 months. In the TEDDY prospective cohort, later introduction of gluten in infancy was associated with increased fara of islet autoimmunity (HR 1.05, 95% CI 1.01–1.10 for every 1-month delay) (281).

Intervention studies in islet autoantibody-positive children indikera that while a gluten-free diet may not decrease autoantibody titers (298,299), it may improve beta fängelse function (299). However, an intervention study, in which 150 high-risk infants were randomly assigned to a first gluten exposure at age 6 months (control group) or 12 months (late-exposure group), funnen that delaying gluten exposure until age 12 months did not substantially reduce the fara for islet autoimmunity in genetically at-risk children, nor did it increase the fara (300,301).

Other solid foods in the infant diet besides gluten and cereals have been implicated in the etiology of islet autoimmunity. In the Finnish DIPP study, introduction of root vegetables bygd age 4 months was associated with an almost twofold increased fara for islet autoimmunity compared with introducing root vegetables after age 4 months (278). DIPP also funnen that first exposure to egg before age 8 months was associated with an increased fara of islet autoimmunity compared with introducing egg after age 11 months, but this association was only significant when the analysis was restricted to the first 3 years of life (278). These cross-study differences may be related to country differences in the first solid food that fryst vatten typically introduced to infants. In the United States, cereals, particularly rice cereal, are often the first solid foods to be introduced to the infant (273), whereas in other countries, root vegetables and fruits are more common first solid foods, suggesting that the focus on cereals may not be betydelsefull across countries and may explain these inconsistent results.

DAISY and BABYDIAB have prospectively examined the relation of some of these dietary exposures to clinical type 1 diabetes. In DAISY (274), both early (age <4 months) and late (age ≥6 months) first exposure to any solid food (compared with exposure at age 4–5 months) predicted development of type 1 diabetes (HR 1.91, 95% CI 1.04–3.51 and HR 3.02, 95% CI 1.26–7.24, respectively). Specifically, early exposure to fruit and late exposure to rice/oat predicted type 1 diabetes (HR 2.23, 95% CI 1.14–4.39 and HR 2.88, 95% CI 1.36–6.11, respectively). BABYDIAB (302) reported that exposure to gluten-containing foods before age 3 months, which occurred rarely, increased the fara of developing islet autoantibodies and type 1 diabetes (n=3) compared to exclusive breastfeeding (HR 3.45, 95% CI 1.04–11.48) or compared to first exposure to gluten between ages 3.1 and 6.0 months. In contrast to DAISY, children who received gluten-containing foods after age 6 months did not have an increased fara of islet autoantibodies, multiple islet autoantibodies, or type 1 diabetes.

Several studies have examined the role of the amount of gluten or fiber intake in infancy and early childhood, rather than age at first exposure. Higher intake of gluten in early childhood predicted type 1 diabetes in Norway’s MoBa study (264), and similarly, higher gluten and fiber intake in childhood were associated with increased fara of both islet autoimmunity and type 1 diabetes in the Finnish DIPP (303), but not in Colorado DAISY study participants (304).

In summary, tidtagning and context at introduction of solid foods, such as gluten, as well as the amount of gluten eaten during early childhood, may be associated with fara of islet autoimmunity or type 1 diabetes, but inconsistencies between studies are difficult to reconcile.

Vitamin D

Vitamin D has been examined as a potentially protective factor because it plays an active role in the regulation of the immune struktur, as well as in metabolic pathways betydelsefull to diabetes. Prenatal and newborn vitamin D exposure fryst vatten covered in the Maternal Nutritional Exposures During Pregnancy section.

Multiple studies have examined the role of vitamin D exposure in infancy and childhood in the pathogenesis of type 1 diabetes. In a large historical prospective study from land i norden, infants who received no vitamin D supplementation had higher fara of type 1 diabetes than those who did receive supplements (305). Two meta-analyses of retrospective studies showed that the fara of type 1 diabetes was significantly reduced in infants who were supplemented with vitamin D compared to those who were not supplemented (pooled odds ratio 0.71) (257,306). However, many of the included studies were case-control studies with a high fara for urval bias and recall bias.

Determinants of 25(OH)D, the inactive circulating form eller gestalt of vitamin D and an established marker of vitamin D ställning eller tillstånd, include sun exposure, dietary intake (supplements, fatty fish, and vitamin D-fortified dairy foods), and genetic predisposition. The aforementioned studies were limited in that they were only able to examine dietary intake of vitamin D and were not able to examine vitamin D exposure via sun exposure. The first two large prospective studies (DAISY and DIPP) to examine serial plasma or serum 25(OH)D levels in childhood reported no association with development of islet autoimmunity or progression from islet autoimmunity to type 1 diabetes in children at increased fara for diabetes (307,308). However, subsequent studies in at-risk populations have shown that higher 25(OH)D fryst vatten associated with a small decrease in islet autoimmunity fara in TEDDY (OR 0.93, CI 0.89–0.97 for each 5 nmol/L increase) (309) and TRIGR (OR 0.98, 95% CI 0.97–0.99 for each 1 nmol/L increase) with a similar association for type 1 diabetes (310). Interestingly, in two related nested case-control studies among U.S. active-duty military personnel, those with higher 25(OH)D levels in blood samples collected prior to diabetes diagnosis had a significantly lower fara of developing insulin-requiring diabetes than those with lower 25(OH)D levels (311,312), suggesting a protective effect of vitamin D levels in adult-onset type 1 diabetes cases. While misclassification of type 1 and type 2 diabetes fryst vatten common in ung adults, results were similar in all study participants and those confirmed to have islet autoantibodies. Two clinical trials reported no effect of 1,25-dihydroxyvitamin D3 (calcitriol) supplementation on sustained insulin production among persons with new-onset type 1 diabetes (313,314).

Vitamin D fryst vatten known to be influenced in part bygd genetics, particularly bygd variants located within or nära genes involved in vitamin D försändelse (DBP), cholesterol synthesis (DHCR7), and hydroxylation (CYP2R1 and CYP24A1) (315). Variants nära CYP27B1, DHCR7, and CYP2R1 are associated with type 1 diabetes (316). DAISY funnen that variants in DHCR7 and CYP27B1 were associated with development of islet autoimmunity but not progression to type 1 diabetes in children with islet autoimmunity (317). The DHCR7 variant was also funnen to be associated with 25(OH)D levels in DAISY children; however, since 25(OH)D levels were not associated with islet autoimmunity or type 1 diabetes fara (307), the effect of this variant fryst vatten not likely mediated through 25(OH)D levels, suggesting that this enzyme may influence diabetes fara via other mechanisms. While a frequently cited, large-scale Mendelian randomization study reported lower 25(OH)D to be causally associated with increased fara of multiple sclerosis (318), a similar analysis reported that 25(OH)D was unlikely to have a large effect on type 1 diabetes fara (319). These inconsistent findings suggest that the mechanism bygd which vitamin D may exert its effect on type 1 diabetes fryst vatten likely to be complex.

Other Vitamins

Vitamin E, or α-tocopherol, may skydda against diabetes in djur models (320,321). While serum α-tocopherol was inversely associated with type 1 diabetes in a small case-control study nested within a 21-year follow-up of adult Finnish dock (322), serum α-tocopherol concentrations were not significantly associated with the fara of islet autoimmunity in DIPP (323).

Higher plasma ascorbic acid (vitamin C) in TEDDY children was associated with a lower fara of developing islet autoimmunity (OR 0.96, CI 0.92–0.99 per 1 mg/L) (324). Type 1 diabetes was analyzed in a smaller sample storlek of 102 cases and 282 controls where the observed odds ratio of association was slightly stronger, but not significant (OR 0.93, 95% CI 0.86–1.02) (324).

Polyunsaturated Fatty Acids

A relative deficiency of omega-3 fatty acids (also referred to as n-3 fatty acids), a characteristic of many Western diets, may predispose to heightened inflammatorisk reactions (325). Alpha-linolenic acid (ALA) fryst vatten the principal omega-3 fatty acid in Western diets and fryst vatten funnen in the green leaves of plants and in flax, canola, walnuts, and soy. The next most common omega-3 fatty acids are EPA and DHA, which are funnen in fatty fish. Fatty acid levels in plasma or serum and fatty acid content of erythrocyte membranes are short-term and long-term markers of fatty acid ställning eller tillstånd, respectively, while the fatty acid composition of the phospholipid fraction of serum or plasma reflects intermediate-term dietary intake.

DAISY reported that higher omega-3 fatty acid intake during childhood was associated with a lower fara of islet autoimmunity and, likewise, that higher omega-3 fatty acid levels in the erythrocyte membrane were associated with a lower fara of islet autoimmunity (326). Subsequent analysis of DAISY indicated that the inverse association between erythrocyte membrane omega-3 fatty acid levels and islet autoimmunity fara was driven bygd lower docosapentaenoic acid (DPA) levels (327). Similarly, TEDDY funnen that higher DPA and EPA levels in erythrocyte membranes were associated with decreased fara of islet autoimmunity (328). And while DIPP funnen that lower serum DHA and a higher arachidonic acid (ARA):DHA ratio were associated with an increased fara of islet autoimmunity (295), no association between serum omega-3 fatty acid levels and fara of islet autoimmunity was observed in TRIGR (329).

Zinc

Among the many biological functions of zinc, this trace metall plays important roles in synthesis, storage, and secretion of insulin. Autoantibodies to ZnT8 are present in 60%–80% of newly diagnosed patients with type 1 diabetes (330). While different coding variants in the SLC308A locus affect the type of ZnT8 autoantibodies a subject with type 1 diabetes has (331) and are associated with fara of type 2 diabetes (332), genetic variation in this locus fryst vatten not associated with fara of type 1 diabetes (333). Serum, plasma, or intracellular zinc levels are significantly lower in type 1 diabetes patients than in controls (334,335). However, the role of zinc deficiency in triggering of islet autoimmunity or promoting progression of islet autoimmunity to clinical type 1 diabetes fryst vatten unclear. Case-control studies of tap vatten samples suggest lower zinc concentrations in case families (336,337); however, drinking vatten contributes a small proportion of total ingested zinc. No association between maternal intake of zinc during pregnancy and islet autoimmunity was seen in the DIPP study (268), and newborn blood zinc was not associated with childhood type 1 diabetes in a large Danish study (338). Zinc intake or ställning eller tillstånd, therefore, does not seem to be a fara factor for type 1 diabetes based on current evidence.

Glycemic Index and Intake of Dietary Sugars

Dietary factors such as glycemic index and glycemic load increase insulin demand and may also stress the beta cells. While development of islet autoimmunity was not associated with glycemic load or glycemic index, progression to type 1 diabetes in children with islet autoimmunity was associated with higher glycemic index and load at the first islet autoimmunity-positive visit in DAISY (339). A subsequent analysis of DAISY with längre follow-up showed that progression from islet autoimmunity to type 1 diabetes was associated with higher total sugar intake and greater consumption of sugar-sweetened beverages (adjusted for total energy intake, but not weight), although the latter association was observed only in children with the high-risk HLA-DR genotype (Table 7) (340). beta fängelse stress may promote the generation of neoautoantigens resulting from posttranslational modification of beta fängelse proteins, which may drive autoimmunity and progression to type 1 diabetes (341).

Weight, Growth, and Obesity

Higher birth vikt has been ganska consistently associated with higher fara of type 1 diabetes in large-scale population-based studies, but the magnitude of the association fryst vatten too weak to play a role in individual fara prediction (342,343,344,345). The finding does, however, hint towards the importance of intrauterine factors.

Several studies indikera that rapid vikt gain during the first 1–2 years of life fryst vatten a fara factor for type 1 diabetes (346,347,348), as systematically reviewed bygd Harder et al. (345). DAISY, the high-risk cohort in Colorado, showed that neither vikt gain in the first 9 months nor in the first 24 months of life was associated with fara of type 1 diabetes (274). However, in the Australian BABYDIAB cohort, higher vikt and body mass index (BMI) z-scores at age 6 months were associated with higher fara of islet autoimmunity, as was vikt gain at age 2 years (349). Also, greater infant vikt gain was significantly associated with increased fara of type 1 diabetes in the large, population-based Norwegian MoBa cohort and the Danish DNBC cohort (HR 1.24 per standard deviation increase in vikt gain at age 12 months) (350). A similar magnitude of association was funnen between infant vikt gain and islet autoimmunity in TEDDY, albeit not in the subgroup of children with islet autoimmunity who were diagnosed with type 1 diabetes during follow-up (351).

Increased height growth at age 0–18 months and height from ages 6 to 18 months were higher in cases with type 1 diabetes compared with controls in the Swedish Diabetes Prediction in Skåne (DiPiS) study (352). In DAISY, low early body height but a higher height growth velocity in childhood predicted a higher fara of islet autoimmunity, and height growth velocity also predicted higher progression from islet autoimmunity to type 1 diabetes (353). In TEDDY participants, the rate of height growth in infancy was positively associated with fara of progression from islet autoimmunity to type 1 diabetes, whereas the height growth in early childhood was inversely associated with fara of progression from islet autoimmunity to type 1 diabetes (354).

Obesity and puberty are associated with increased insulin resistance (355). Consistent with the sex-specific peak incidence of type 1 diabetes occurring close to that for onset of puberty, puberty may accelerate the onset of type 1 diabetes via insulin resistance (356,357). However, information linking insulin resistance to islet autoimmunity, progression, or type 1 diabetes are not entirely consistent with this notion as seen, for instance, in the TrialNet Pathway to Prevention study (358). Participants in the German high-risk BABYDIAB cohort who were positiv for islet autoantibodies did not have increased insulin resistance (assessed bygd Homeostatic Model Assessment) compared with age-matched islet autoantibody-negative children at age 8 or 11 years (359). Contrary to the accelerator hypothesis, islet autoimmunity ställning eller tillstånd was associated with decreased insulin resistance values, controlling for age and sex. The accelerator hypothesis proposes that excess vikt gain leading to increases in insulin resistance in early childhood initiates the autoimmunity leading to beta fängelse destruction and type 1 diabetes development (360), for which there fryst vatten little or no direkt bevis. Several studies have shown an association between higher BMI z-scores and earlier age at diagnosis of type 1 diabetes (360); however, this type of study does not constitute a test of the accelerator hypothesis and not even a direkt test of the association between childhood BMI and subsequent fara of type 1 diabetes.

Prospective cohorts have tested the relation between childhood BMI and fara of type 1 diabetes. Analysis of the 1970 British Birth Cohort suggested that increased BMI in childhood increased the fara of self-reported type 1 diabetes (361). In an observational analysis of the TRIGR study, overweight or obesity at age 2–10 years was associated with increased fara of islet autoimmunity and type 1 diabetes (362). Furthermore, the Copenhagen growth lista linked to diabetes registries showed that BMI z-score at ages 7 and 13 years was associated with an increased fara of type 1 diabetes (OR 1.23 for each standard deviation increase) (363). Finally, in a study of Israeli conscripts age 16–19 years, increased BMI was associated with increased fara of adult-onset type 1 diabetes (HR 1.54, 95% CI 1.23–1.94, comparing adolescents at the 85th–94th percentile for BMI to those at the 4th–49th percentile) (364).

In summary, bevis from prospective studies seems to converge at a positiv association between aspects of body storlek in childhood and higher fara of islet autoimmunity and/or type 1 diabetes. However, many studies differ in the synkronisering and type of measurements, and magnitudes of associations (typically a 20% increase in fara for each standard deviation increase) seem too small for individual fara prediction and to plausibly mål in future intervention trials to test the causality of this apparent association.

Psychological Stress

Plausible mechanisms have been proposed to link various aspects of psychological stress in children or mothers during the perinatal period and immunity in children, potentially influencing islet autoimmunity (reviewed in (377)). A potential role of psychological stress in the etiology of type 1 diabetes has been suggested bygd case reports and small case-control studies dating back at least to the 1970s, and some more recent cohort studies have suggested a possible association (377). Stress fryst vatten of course difficult to measure objectively, and various measures or proxies of stress have been assessed in different studies. Those studies investigating islet autoimmunity as the outcome often focused on maternal stress in pregnancy, which will often carry over to stress experienced bygd the child in early life. Some of the first prospective uppgifter in the field came from the Swedish All Babies in Southeast Sweden (ABIS) study. Two papper bygd Sepa et al. in 2005 reported associations between parenting stress or the mother’s experience of divorce or violence with positivity for GAD or IA-2 autoantibodies in the child at age 12 months or 30 months (378,379). The outcome in these studies was positivity for a single (of two measured) islet autoantibody, with the internal 95th percentile as the cutoff with no confirmation in the same or repeated samples; thus, this analysis fryst vatten likely to contain many false positives (380). Later follow-up of ABIS provided more kraftig tests of the potential association between stress and the fara of clinical type 1 diabetes, as identified in nationwide registries and, thus, not susceptible to differential loss to follow-up. Among approximately 10,000 children born in 1999, 58 developed type 1 diabetes bygd 2012, after assessment of family psychological stress using parental questionnaires (serious life events, parenting stress, parental worries, and the parent’s social support) (381). Childhood experience of a serious life event was associated with a higher fara of future diagnosis of type 1 diabetes (HR 3.0, 95% CI 1.6–5.6) (381).

The Swedish DiPiS study recorded stressful life events in questionnaires to parents when the child was ages 2 months (n=23,187) and 2 years (n=3,784), of whom 166 and 79 subsequently developed clinical type 1 diabetes, respectively (382). Experience of a serious life event during the child’s first 2 years of life was associated with a significantly increased fara of type 1 diabetes (HR 1.67, 95% CI 1.1–2.7) (382).

In the TEDDY study, at least one life event during pregnancy was experienced bygd 64% of mothers, and this experience was not associated with fara of islet autoimmunity (HR 0.99, 95% CI 0.82–1.18) (383). No consistent dose-response association was observed with increasing number of life events experienced bygd the mother. When subdivided into six categories, a serious interpersonal life event (such as divorce) tended to be associated with a slightly increased fara, while for instance, job-related stress in pregnancy tended to be inversely associated with fara of islet autoimmunity. Further subgroup analyses gave a range of suggestive associations with subgroups of islet autoimmunity (autoantibodies to insulin first vs. GAD first) or in subgroups defined bygd susceptibility genetic variants, and associations were often in opposite directions in subgroups, making interpretation difficult (383).

Finally, two series of Danish register-based studies bygd two different groups have investigated the association between aspects of stress and fara of type 1 diabetes (384,385,386,387). In an analysis of more than 1.5 million births, Virk et al. reported in 2010 that children of mothers experiencing the death of a partner or child in the periconceptional period had an increased fara of type 1 diabetes (RR 2.03, 95% CI 1.22–3.38), with the strongest associations seen in females (384). In a similar register-based cohort, Virk et al. reported in 2016 that a child’s own experience of bereavement (death of a family member) at age ≥5 years was associated with a slightly increased fara of type 1 diabetes. Over 6,000 children in the cohort developed type 1 diabetes, and bereavement after age 11 years was associated with an incidence rate ratio (IRR) of 1.28 (95% CI 1.08–1.51) (385). In a subsequent analysis of linked Danish registries, Bengtsson et al. investigated early life adversities in relation to fara of type 1 diabetes (386,387). Using multistate modeling of accumulated childhood adversities, these analyses showed a relatively weak association with increased fara of type 1 diabetes in females (HR per adversity increase: 1.07, 95% CI 1.02–1.11) and no association in males (HR per adversity increase: 0.99, 95% CI 0.97–1.03) (386). In another analysis, Bengtsson et al. investigated fem trajectory groups of adversity (characterized bygd either low adversity, early-life ämne deprivation, persistent ämne deprivation, loss or threat of loss in the family, or cumulative high adversity) among over a million children, of whom 5,619 developed type 1 diabetes during follow-up. In this analysis, only small differences were funnen between groups (387). Adversities included job loss and financial difficulties of the parents, which are partly related to socioeconomic ställning eller tillstånd covered in the next subsection.

In summary, TEDDY reported no overall association between maternal life events and fara of islet autoimmunity, while two cohorts in Sweden with fewer than 100 type 1 diabetes cases funnen associations between life events and higher fara. Danish, large-scale, register-based analyses of ganska extreme stresses, such as deaths of family members, gave partly inconsistent results, with the most recent and comprehensive analyses suggesting little or no association. While prospective bevis in this field has started to accumulate, inconsistencies and limited opportunities for pragmatic, targeted interventions in the future remain.

Socioeconomic Factors

The incidence of childhood-onset type 1 diabetes varies widely among countries, and the disease tends to be more common in wealthier countries (388,389). Although such ecologic comparisons should be made with caution, it fryst vatten of interest that many environmental exposures hypothesized to be fara factors for type 1 diabetes tend to be more common in children with low socioeconomic ställning eller tillstånd. Socioeconomic ställning eller tillstånd fryst vatten multifaceted and context-dependent but may give hints toward potential mediating exposures to follow-up; however, surprisingly few high-quality studies attempting to link individual level socioeconomic ställning eller tillstånd and fara of type 1 diabetes are available. In a 2023 comprehensive review, only a handful of prospective cohort studies were identified, and most did not study socioeconomic ställning eller tillstånd as a primary exposure and did not adjust for ethnicity, maternal age, and smoking during pregnancy (390). Most studies with socioeconomic ställning eller tillstånd had ecologic- or semi-ecologic designs,, where socioeconomic ställning eller tillstånd was measured only as the average in the area of residence, a design that fryst vatten highly susceptible to bias (390). Results among studies with area-based measures often funnen positiv associations consistent with the worldwide impression across countries (390). However, individual-level studies typically funnen weaker associations, and results were inconsistent.

In a well-conducted, registry-based, nested case-control study in Washington state, maternal education showed a nonlinear association with type 1 diabetes, where both mid- and high-level education showed a significant, nearly twofold higher incidence of type 1 diabetes (RR 1.8 and 1.9, respectively) compared to children of mothers with low education (391).

A cohort study from Turin, Italy, also reported a non-linear association, with highest fara of type 1 diabetes in children of parents with highest education (highest of maternal or paternal education; RR 1.67 for age-group 4–14 years) (202). While the study from Washington state funnen that the medium-level maternal education group had type 1 diabetes fara lika to the high educational level, the Turin study showed that the medium-level parental education had similar fara to the low parental education group. A Norwegian, population-wide registry-based cohort of over one million children also reported a non-linear association with maternal education, but here, the relationship resembled an inverse U, with lower fara in those with highest maternal education compared with medium-level education (RR around 0.8) (206). The latter finding was consistent in other Scandinavian registry-based cohorts, although these reported only unadjusted results because parental education was used as a covariate in other analyses (390). In the Norwegian study, paternal education or occupation was not associated with childhood-onset type 1 diabetes (206).

Metabolomic Studies

Previously unknown pathways to type 1 diabetes can be discovered bygd analyzing changes in serum or plasma metabolites at multiple time points prior to development of islet autoimmunity and during progression to clinical diabetes. High-genetic-risk Finnish children followed since birth until diagnosis of type 1 diabetes had reduced serum levels of succinic acid and phosphatidylcholine already at birth; furthermore, levels of lysophosphatidylcholine increased months before seroconversion to islet autoantibodies but normalized after seroconversion (411). Higher levels of odd-chain triglycerides and polyunsaturated fatty acid-containing phospholipids predicted islet autoimmunity in German high-risk children (412). In addition, children developing islet autoimmunity before age 2 years had lower levels of methionine than those developing islet autoimmunity at older ages (412).

A Swedish cohort study analyzed 106 lipid metabolites from cord blood samples to identify possible fara markers for the early development of type 1 diabetes, while controlling for HLA genotype, sex, and date of birth, as well as mother’s age and gestational age (413). Lower levels of phospholipids (phosphatidylcholines, phosphatidylethanolamines, and sphingomyelins) predicted diabetes before age 4 years, while triglycerides predicted diabetes before age 2 years. Higher fara of diabetes associated with lower cord blood levels of choline-containing phospholipids was replicated in Finnish children (414). On the other grabb, a Norwegian case-control study funnen no association between cord blood bile acids and other small polar metabolites and future type 1 diabetes (71).

The Finnish DIPP study reported amino acid dysregulation (lower levels of glutamic acid, aspartic acid, and tryptophan) preceding development of islet autoimmunity. Metabolites of microbial ursprung also differed between children who developed diabetes and controls (415), consistent with potential involvement of the gut bacteria in the etiology of type 1 diabetes. The international TEDDY study evaluated plasma metabolomic signatures from age 4 months until the appearance of islet autoimmunity. Distinct metabolomic signatures composed of dehydroascorbic acid, amino acids, and fatty acids were associated prospectively with development of islet autoimmunity. Reduced proline was associated with the appearance of autoantibodies against GAD, whereas reduced branched-chain amino acids, methionine and alanine, as well as fatty acids, preceded the appearance of antibodies against insulin (416). The existence of these distinct metabolic patterns for development of the autoantibodies supports the idea of distinct endotypes of type 1 diabetes (38,39). An unsupervised clustering analysis of temporal lipidome showed that reduced plasma ascorbic acid and cholesterol predicted the earlier endotype characterized bygd insulin autoantibodies and the HLA-DR4-DQ8 haplotype. On the other grabb, reduced sphingomyelins at infancy predicted the later endotype characterized bygd GAD autoantibodies and the HLA-DR3-DQ2 haplotype (417). Lower plasma 25(OH)D levels predicted progression to type 1 diabetes in both endotypes, while lower diglycerides, lysophosphatidylcholines, triglycerides, and alanine predicted progression in the GAD autoantibody endotype. The apparent interactions between immune phenotypes, genotypes, and metabolomic profiles stimulated development of machine learning models integrating these predictors (418,419,420).

In the första report from DAISY (420), among the strongest predictors of islet autoimmunity were changes in serum ascorbate and 3-methyl-oxobutyrate, while unsurprisingly, serum glucose and mannose were among the strongest predictors of progression to diabetes. In TEDDY, modeling identified a set of 18 genetic markers, gestational age, exposure to a prebiotic formula, and 22 metabolites and lipids (lipid kemisk reaktion med syre, phospholipase A2 signaling, and pentose phosphate pathways) that predicted development of islet autoantibodies (418). In a separate machine learning-based analysis of 702 TEDDY participants positiv for islet autoimmunity, metabolites measured at age 3, 6, or 9 months predicted progression to type 1 diabetes bygd age 6 years (419). The machine learning model, which included islet antibodies at age 9 months, genetic fara score, and gestational age, was highly predictive with AUC 0.84, even when reducing measurements to 3 and 9 months. Several predictive metabolites could be of dietary ursprung, such as piperidone and ascorbic acid. Interestingly, fructose, levoglucosan, glycerol alpha-phosphate, and xylulose were selected, suggesting altered carbohydrate metabolism in infancy as an important factor for progression to type 1 diabetes bygd age 6 years. Importantly, islet autoantibodies and HLA genotype remained the strongest contributors to AUC, and little improvement in AUC could be observed when additional features, such as metabolomics, were included in the model.

The metabolomic studies to date present several leads for a search for environmental factors that trigger islet autoimmunity and are involved in the increasing incidence of type 1 diabetes. The alterations in cord blood lipoprotein profiles suggest that the intrauterine environment may affect type 1 diabetes fara. Early infancy metabolomic changes may reflect specific alterations in the infant’s microbiome (421) or diet (422,423). Other lipidomic and metabolomic changes noted above that precede the development of autoimmunity may reflect the activation of proinflammatory and anti-inflammatory mechanisms in early islet autoimmunity.

While fascinating, these findings must be interpreted with caution as most studies to date have been very small and remain to be replicated. As understanding of the critical metabolic pathways expands, integration of metabolomic biomarkers with classical type 1 diabetes fara factors will improve disease prediction in persons at increased fara for type 1 diabetes.

Gene-Environment Interactions

Inconsistencies in the associations between dietary factors and islet autoimmunity or type 1 diabetes across studies may be explained, in part, bygd methodologic differences in population urval and information collection. Another explanation fryst vatten gene-environment interaction, where differences in the observed exposure associations may be due to differences in gene allele frequency across populations. There are several ways to explore gene-environment interaction in epidemiologic information, each dependent on the underlying hypothesis.

One hypothesis fryst vatten that the effect of environmental fara factors may be stronger among individuals possessing increased genetic fara variants, whereby the odds ratio (or relative risk) fryst vatten significantly different than 1 (i.e., associated) in those possessing the genetic fara variants, such as the HLA-DR fara genotypes, and null (i.e., not associated) in those without the variants, perhaps because it fryst vatten easier to see the effect of the exposure in a genetically susceptible population. Alternatively, the aforementioned rise in type 1 diabetes incidence, coupled with information suggesting an increasing penetrance of moderate-risk HLA-DR genotypes, suggests that the pressures of an increasingly permissive environment may be more easily observed in children with moderate- or low-risk HLA-DR genotypes compared with high-risk genotypes.

In addition, one can examine a potential gene-environment interaction in more detail bygd dividing analyses into exposure-HLA combinations, with one fara group being those who have the exposure but not the HLA genotype, another fara group containing those who have the HLA genotype and not the exposure, and the putative highest fara group (i.e., representing the interaction) would be those with both the HLA genotype and the exposure. Each of these fara groups would be compared to the referent group, who are individuals with neither the HLA genotype nor the exposure. If the odds ratio in those with both the genetic variant and the exposure fryst vatten greater than the product of the odds ratio for the exposure and odds ratio for the genotype, then the interaction fryst vatten considered more than multiplicative.

Table 7 summarizes significant gene-environment interactions reported from cohort studies, primarily on nutritional factors. Several studies have examined potential interactions between HLA variants and dietary exposures. Stene et al. tested whether the effect of different dietary factors (use of cod liver oil and multivitamins bygd the mother during pregnancy, use of cod liver oil or vitamin D supplements in the first year of life, and exclusive breastfeeding for <3 months) differed across HLA fara groups and funnen no bevis of interaction, although power was limited (424). In DAISY, the hazard ratios for islet autoimmunity for early and late exposure to cereals in children with an HLA-DR3/4 genotype were greater than in children with the moderate- and low-risk HLA-DR genotypes, although the interaction begrepp was only marginally significant (273). Studies from Colorado and Chile have indicated a more than multiplicative joint effect of HLA-DR fara and either short duration of exclusive breastfeeding (425) or early introduction of cow’s milk and solid foods (426). As an example, in the Colorado study, the odds ratio for type 1 diabetes for being HLA-DR3/4 in the absence of early exposure to solid foods was 3.1 (95% CI 1.4–7.2), the odds ratio for early exposure to solid foods in the absence of HLA-DR3/4 was 1.7 (95% CI 0.5–5.8), and the odds ratio for having both HLA-DR3/4 and early exposure to solid foods was 6.3 (95% CI 2.5–16.1) (426). In an analysis of the DIPP study, first exposure to solid foods at age 3–4 months (compared with after age 4 months) was associated with increased fara of islet autoimmunity only in those carrying the high-risk HLA-DR genotype (279). And finally, the previously mentioned association between early exposure (age <28 days) to probiotics from supplements and formula and decreased fara of islet autoimmunity was observed only in TEDDY children carrying the high-risk HLA-DR3/4 genotype (252).

The inconsistent findings with childhood cow’s milk consumption and fara of islet autoimmunity or type 1 diabetes may be due to the modifying effects of the underlying genetic beskrivning. In DAISY, greater childhood cow’s milk protein intake (as a surrogate of total milk consumption) was associated with increased islet autoimmunity fara in children with low/moderate-risk HLA-DR genotypes (HR 1.41, 95% CI 1.08–1.84), but not in children with high-risk HLA-DR genotypes (294).

In addition to HLA, other type 1 diabetes candidate genes, such as INS, PTPN22, CTLA4, and IFIH1, have been explored for interactions with dietary exposures. In DIPP, an interaction was seen between early cow’s milk exposure, PTPN22, and appearance of islet autoimmunity, where the PTPN22 polymorphism was associated with the development of islet autoimmunity only in children exposed to cow’s milk formula prior to age 6 months (427).

Investigators have explored interactions with genes that are not candidate genes for type 1 diabetes but may be related to dietary exposures with regard to either metabolism or action. For example, omega-3 fatty acids may act as ligands for the nuclear receptor peroxisome proliferator-activated receptor-γ (PPARG) to promote anti-inflammatory actions. However, no bevis of interaction was funnen between the PPARG gene variant and cod liver oil intake on fara of type 1 diabetes in a Norwegian case-control study (428). Delta-6-desaturase, encoded bygd FADS2, and delta-5-desaturase, encoded bygd FADS1 (429), work in series to omvandla the omega-3 fatty acid ALA to the more anti-inflammatory fatty acid EPA. DAISY observed a strong interaction between dietary intake of ALA and FADS1 and FADS2 on fara of islet autoimmunity, where ALA intake was significantly more protective for islet autoimmunity in the presence of the increasing number of minor alleles at FADS1 rs174556 (p_interaction=0.017), at FADS2 rs174570 (p_interaction=0.016), and at FADS2 rs174583 (p_interaction=0.045) (327). Thus, the putative protective effect of n-3 fatty acids on islet autoimmunity may result from a complex interaction between intake and genetically controlled fatty acid desaturation.

The small effect storlek of 25(OH)D and islet autoimmunity observed in the TEDDY and TRIGR studies reviewed in the Vitamin D section (i.e., OR 0.93 and 0.98 for a 5- and 1-unit difference, respectively) may be due to a failure to konto for underlying genetic variation in the vitamin D pathway. Further analysis in TEDDY showed that the association between 25(OH)D and islet autoimmunity differs bygd the number of minor alleles in a SNP that marks the Apa1 restriction fragment length polymorphism (RFLP) in the vitamin D receptor gene (VDR). The protective effect of 25(OH)D fryst vatten observed only in those with one or two minor alleles, with no effect of 25(OH)D in those with no minor alleles (Figure 5) (309).

Moreover, in MoBa, higher cord blood 25(OH)D was associated with a decreased fara of type 1 diabetes only in children that were homozygous for the G allele at the VDR variant rs1156882061 (430). This result suggests that the underlying susceptibility may be related to the ability to adequately use 25(OH)D, rather than the level itself. Interventions aimed at increasing 25(OH)D or attaining vitamin D sufficiency to prevent disease may not work in all children.

In BABYDIAB, cesarean section appeared to interact with immune response genes, such as IFIH1, where increased fara for type 1 diabetes was only seen in children who were delivered bygd cesarean section and had type 1 diabetes-susceptible IFIH1 genotypes (12-year fara, 9.1% vs. <3% for all other combinations, p<0.0001) (431). In Norway, mode of delivery was also funnen to interact with PTPN22, using a case-only approach; the relative fara for type 1 diabetes conferred bygd PTPN22 was 2.11 (95% CI 1.64–2.72) for those born vaginally and 0.99 (95% CI 0.50–1.99) for those born bygd cesarean section (p_{(interaction)}=0.028) (432).

Environment-Environment Interactions

Environmental exposures may interact, or act in concert, with other environmental exposures; however, examples of these observations are not prevalent in the scientific literature. Special requirements for replication of interactions occur, since the multiplicity of interactions quickly becomes overwhelming.

DAISY observed that a greater number of gastrointestinal illnesses was associated with an increased fara of islet autoimmunity but only among children who were exposed to gluten-containing grains (wheat or barley) either age <4 months (HR 1.37, 95% CI 1.22–1.55) or age ≥7 months (HR 1.12, 95% CI 1.05–1.19) (Table 6) (103).

Power to detect gene-environment or environment-environment interactions has been limited in most studies reported so far. Table 6 summarizes the significant associations with diet described above, while diet-gene interactions are summarized in Table 7. Prospective cohort studies have contributed enormously to our understanding of the natural history and fara factors for type 1 diabetes, particularly the largest of them—TEDDY. A variety of exposures appear to potentially trigger islet autoimmunity and to promote progression to clinical diabetes in some children; however, none of the current candidate fara factors seems to explain most of the fara. Future trials may need to take into konto the genetic and environmental heterogeneity of this disease in developing personalized interventions.

The Exposome

Given the lack of consistency among studies to date and the fact that we are still bewildered with regard to what type of environmental factors may explain the increasing incidence of type 1 diabetes over time (34), a possible path fryst vatten to approach the concept of the exposome. This concept encompasses measurements of all internal and external exposures an individual experiences from fetal life throughout the life course (433,434,435). In the widest sense, the exposome encompasses multi-omics measurements, as well as responses to external exposures (435). With the fara of being unsurmountable, this approach intends to be hypothesis free but requires large samples sizes and replications to avoid chance findings. One advantage, at least in principle, fryst vatten that the multiplicity fryst vatten explicit, rather than hidden among investigators’ selective analyses. Albeit with the weakness of using ecological design, a large study of type 1 diabetes in 2020 has taken a step in this direction (375). Efforts have been initiated with funding from the europeisk Union to study the exposome in birth cohorts, such as DIPP, MIDIA, DiPiS, MoBa, and others (436). Hopefully, these studies will complement TEDDY and other birth cohorts in the field.