Neurobehavioral decision profiles have often been neglected in chronic diseases despite their direct impact on major public health issues such as treatment adherence. This remains a major concern in diabetes, despite intensive efforts and public awareness initiatives regarding its complications. We hypothesized that high rates of low adherence are related to risk-taking profiles associated with decision-making phenotypes. If this hypothesis is correct, it should be possible to define these endophenotypes independently based both on dynamic measures of metabolic control (HbA1C) and multidimensional behavioral profiles.

In this study, 91 participants with early-stage type 1 diabetes fulfilled a battery of self-reported real-world risk behaviors and they performed an experimental task, the Balloon Analogue Risk Task (BART).

K-means and two-step cluster analysis suggest a two-cluster solution providing information of distinct decision profiles (concerning multiple domains of risk-taking behavior) which almost perfectly match the biological partition, based on the division between stable or improving metabolic control (MC, N = 49) v. unstably high or deteriorating states (NoMC, N = 42). This surprising dichotomy of behavioral phenotypes predicted by the dynamics of HbA1C was further corroborated by standard statistical testing. Finally, the BART game enabled to identify groups differences in feedback learning and consequent behavioral choices under ambiguity, showing distinct group choice behavioral patterns.

These findings suggest that distinct biobehavioral endophenotypes can be related to the success of metabolic control. These findings also have strong implications for programs to improve patient adherence, directly addressing risk-taking profiles.

尽管神经行为决策对治疗依从性等主要公共卫生问题有直接影响，但在慢性疾病中常常被忽视。尽管针对糖尿病并发症做出了大量努力并采取了公众意识举措，但这仍然是糖尿病的一个主要问题。我们假设高低依从率与决策表型相关的冒险行为有关。如果这个假设是正确的，那么应该可以根据代谢控制 (HbA1C) 的动态测量和多维行为特征独立地定义这些内表型。

在这项研究中，91 名患有早期 1 型糖尿病的参与者完成了一系列自我报告的现实世界风险行为，并执行了一项实验任务，即气球模拟风险任务 (BART)。

K均值和两步聚类分析建议采用两聚类解决方案，提供不同决策概况的信息（涉及冒险行为的多个领域），该解决方案几乎完全匹配生物分区，基于稳定或改善代谢控制之间的划分（ MC， N = 49) v.不稳定的高或恶化状态 (NoMC， N = 42)。标准统计测试进一步证实了 HbA1C 动态预测的这种令人惊讶的行为表型二分法。最后，BART 游戏能够识别反馈学习中的群体差异以及模糊情况下随后的行为选择，显示出不同的群体选择行为模式。

这些发现表明，不同的生物行为内表型可能与代谢控制的成功有关。这些发现对于提高患者依从性、直接解决风险承担问题的计划也具有重要意义。