No abstract

To combine prospective cohort studies, by including HLA harmonization, and estimate risk of islet autoimmunity and progression to clinical diabetes. RESEARCH DESIGN AND METHODSFor prospective cohorts in Finland, Germany, Sweden, and the U.S., 24,662 children at increased genetic risk for development of islet autoantibodies and type 1 diabetes have been followed. Following harmonization, the outcomes were analyzed in 16,709 infants-toddlers enrolled by age 2.5 years. RESULTSIn the infant-toddler cohort, 1,413 (8.5%) developed at least one autoantibody confirmed at two or more consecutive visits (seroconversion), 865 (5%) developed multiple autoantibodies, and 655 (4%) progressed to diabetes. The 15-year cumulative incidence of diabetes varied in children with one, two, or three autoantibodies at seroconversion: 45% (95% CI 40-52), 85% (78-90), and 92% (85-97), respectively. Among those with a single autoantibody, status 2 years after seroconversion predicted diabetes risk: 12% (10-25) if reverting to autoantibody negative, 30% (20-40) if retaining a single autoantibody, and 82% (80-95) if developing multiple autoantibodies. HLA-DR-DQ affected the risk of confirmed seroconversion and progression to diabetes in children with stable single-autoantibody status. Their 15-year diabetes incidence for higher-versus lower-risk genotypes was 40% (28-50) vs. 12% . The rate of progression to diabetes was inversely related to age at development of multiple autoantibodies, ranging from 20% per year to 6% per year in children developing multipositivity in #2 years or >7.4 years, respectively. CONCLUSIONSThe number of islet autoantibodies at seroconversion reliably predicts 15-year type 1 diabetes risk. In children retaining a single autoantibody, HLA-DR-DQ genotypes can further refine risk of progression.

We propose new privacy attacks to infer attributes (e.g., locations, occupations, and interests) of online social network users. Our attacks leverage seemingly innocent user information that is publicly available in online social networks to infer missing attributes of targeted users. Given the increasing availability of (seemingly innocent) user information online, our results have serious implications for Internet privacy—private attributes can be inferred from users’ publicly available data unless we take steps to protect users from such inference attacks. To infer attributes of a targeted user, existing inference attacks leverage either the user’s publicly available social friends or the user’s behavioral records (e.g., the web pages that the user has liked on Facebook, the apps that the user has reviewed on Google Play), but not both. As we will show, such inference attacks achieve limited success rates. However, the problem becomes qualitatively different if we consider both social friends and behavioral records. To address this challenge, we develop a novel model to integrate social friends and behavioral records, and design new attacks based on our model. We theoretically and experimentally demonstrate the effectiveness of our attacks. For instance, we observe that, in a real-world large-scale dataset with 1.1 million users, our attack can correctly infer the cities a user lived in for 57% of the users; via confidence estimation , we are able to increase the attack success rate to over 90% if the attacker selectively attacks half of the users. Moreover, we show that our attack can correctly infer attributes for significantly more users than previous attacks.