Diabetes In Somali Children In Minneapolis and St.Paul, Minnesota

Abstract

Minnesota is home to the largest population of Somalis in the United States (US). The 2011 United States Census Bureau estimates that there are more than 32,000 individuals of Somali origin [1, 2], and many believe there is a significant underestimation of this number. Somalis originate from a distinct geographic region in East Africa, and have obvious similarities in physical features suggesting relative genetic homogeneity. In contrast, most African Americans descend from various areas in Africa, and have experienced significant intermingling amongst themselves and with other populations rendering this population less homogeneous. Thus, not surprisingly, diabetes may differ between Somalis and other African Americans. Type 1 diabetes (T1D) is an autoimmune disease of heterogeneous etiology [3-5]. Carrying specific human leukocyte antigen (HLA) alleles determines an individual’s genetic risk for developing T1D. It has been established that these alleles may differ between populations [3, 6]. This dissertation describes some of the clinical characteristics, diabetes autoantibody and HLA allele profiles, and explores cultural beliefs related to diabetes in Somali children with T1D who live in the Twin Cities, Minnesota. These studies have led to two papers, which have been submitted for peer-reviewed publication. My primary project, describes the immune and genetic basis of T1D in a group of Somali children in the Twin Cities, Minnesota by determining human leukocyte antigen (HLA) alleles and diabetes autoantibodies. Twenty-seven Somali children ≤19 years treated for T1D at the University of Minnesota and Children’s Hospitals and Clinics of Minnesota from January 1st, 2012 to January 31st, 2014, participated. Venous blood samples for HLA alleles, and diabetes autoantibodies (GAD65, islet antibodies, insulin antibodies and ZnT8) were obtained. In these 27 children, the most common HLA phenotype was DR3. Strikingly, 92% of subjects carried this phenotype (allele frequency 63%). Another common genotype was DR13 (27%, allele frequency 14%). There was a relatively low frequency of DR4 (15%, allele frequency 8%). This genetic pattern is very different from that of Caucasians or African Americans. All 27 participants had positive elevation of at least one diabetes autoantibody confirming that this is autoimmune diabetes. GAD65 antibodies were found in 56% of subjects, IA-2 in 33%, and ZnT8 in 22%. HASH(0x7f87dd84a058) My second project was a cross-sectional study that describes cultural beliefs related to diabetes in Minnesota Somali children with T1D, and compares their diabetes control to that of non-Somali children with T1D. Demographic and clinical data were collected by history and from medical records of Somali children ≤19 years with T1D followed at the University of Minnesota Children’s Hospital and Children’s Hospitals and Clinics of Minnesota. A survey was administered to parents of all participants and to children aged ≥12 years. Twenty-five Somali children participated, with 24 parent-child pairs (2 siblings). In general, diabetes was well accepted. Seventy-one percent of parents indicated the child was “the same as before” other than having to do diabetes cares. Families were coping well, and the child was not treated differently than siblings. Performance of routine medical cares was described as the hardest part about having diabetes, but this was not related to conflicts with traditional culture or religion. One notable exception was difficulty performing carbohydrate counting on Somali foods. Our education materials were not helpful when it came to the traditional Somali diet. Respondents were appreciative of the education provided by the diabetes team. Less than 10% used herbal supplements in addition to insulin. Mean HbA1c in Somali children was higher than the overall pediatric clinic average, 9.5±1.6% vs 8.8±1.6 (p =0.01). The difference was largely due to adolescent patients. These two studies suggest that autoantibody and HLA profiles of Somali children with diabetes are consistent with autoimmune T1D, and that their HLA profile is unique compared to African Americans with T1D. The data also highlight that the majority of Somali families cope well with diabetes, and that glycemic control in adolescents is worse than that of non-Somali peers. Findings from this study beg the question of whether differences in diabetes between Somali children and their non-Somali peers is merely a result of these differences in HLA alleles, or whether other factors influence glycemic control in this population. Data from this study will be used to target diabetes education and to provide culture-specific educational resources to improve the experience of living with this chronic condition.