Diaceutics Controlled Diabetes Dataset

The Diaceutics Controlled Diabetes Dataset offers structured, de-identified real-world data sourced from over 600 U.S.-based laboratories, focusing on diagnostic activity related to diabetes management. This dataset specifically identifies patients with controlled diabetes, defined by Hemoglobin A1c (HbA1c) levels between 6.5% and 7.0%, a clinically recognized range indicating managed glycemic control.

Each record represents a unique laboratory test event, with the TEST_DATE field capturing when results were reported. Test-level data is standardized using LOINC codes, and includes fields such as RESULT_NAME, RESULT_VALUE, RESULT_REFERENCE_RANGE, and RESULT_UNITS_OF_MEASURE.

Patient-level information includes a hashed PATIENT_ID and GENDER, while the ordering physician is identified via PHYSICIAN_NPI. This structure enables cohort definition and analysis based on test name, LOINC, result thresholds (e.g., HbA1c 6.5-7.0%), physician, gender, and year of birth.

Diaceutics supports advanced linkage capabilities via Datavant tokens, enabling integration with diagnosis codes or therapy exposures for enhanced stratification in real-world evidence (RWE) and health economics and outcomes research (HEOR) applications. Common use cases include:

- Diagnostic access planning
- Trial feasibility analysis
- Patient segmentation based on test-defined criteria

Data is delivered through the Snowflake Marketplace under a contract-based access model, with flexible delivery frequencies, including weekly updates.

HbA1c is a critical biomarker for assessing long-term glucose regulation. It reflects the average blood glucose concentration over the prior 2-3 months by measuring the percentage of glycated hemoglobin in red blood cells. This makes it a more stable indicator of glycemic control than daily glucose readings.

In clinical practice, HbA1c is used to:

- Diagnose diabetes and prediabetes
- Monitor the effectiveness of diabetes management strategies
- Evaluate risk for long-term complications such as neuropathy, nephropathy, and retinopathy

原始信息汇总

Diaceutics Controlled Diabetes Dataset

数据集概述

包含过去12个月内受控糖尿病检测事件相关的实验室数据记录。该数据集提供结构化、去标识化的真实世界数据，来源于美国600多家实验室，重点关注糖尿病管理相关的诊断活动。

数据特征

• 患者定义：识别糖化血红蛋白（HbA1c）水平在6.5%至7.0%之间的受控糖尿病患者
• 时间范围：过去12个月的数据
• 地理覆盖：美国所有州
• 更新频率：每周更新

数据内容

患者信息

• PATIENT_ID：哈希处理的患者标识符
• GENDER：性别
• YEAR_OF_BIRTH：出生年份

医生信息

• PHYSICIAN_NPI：医生NPI标识符
• PHYSICIAN_SPECIALTY：医生专业

检测信息

• TEST_ID：检测标识符
• TEST_DATE：检测日期（结果报告日期）
• LOINC_NAME：LOINC名称
• LOINC_CODE：LOINC代码
• TEST_NAME：检测名称

结果信息

• RESULT_ID：结果标识符
• RESULT_NAME：结果名称
• RESULT_VALUE：结果数值
• RESULT_REFERENCE_RANGE：结果参考范围
• RESULT_UNITS_OF_MEASURE：结果测量单位

业务应用场景

受众激活

基于实时实验室检测行为定位医疗专业人员，识别订购关键检测的医生，在患者符合诊断标准时激活外展活动。

需求预测

按检测、生物标志物或地理位置跟踪真实世界诊断需求，分析600多个实验室的检测订购量趋势。

生命科学商业化

利用基于诊断的洞察加速上市规划和市场执行，了解检测可及性、数量和差距。

患者360视图

访问去标识化、以实验室为基础的患者视图，包含检测结果、诊断标志和纵向背景。

市场分析

分析各地区和提供商的诊断市场份额、检测采用率和实验室覆盖范围。

真实世界数据

解锁专注于实验室结果和生物标志物趋势的精选真实世界数据。

技术特性

• 使用LOINC代码进行检测级数据标准化
• 支持通过Datavant令牌进行高级链接功能
• 可与诊断代码或治疗暴露整合
• 支持基于检测名称、LOINC、结果阈值、医生、性别和出生年份的队列定义和分析

交付方式

通过Snowflake Marketplace在基于合同的访问模式下交付，提供灵活的交付频率。