Diaceutics Complete Blood Count Dataset

The Diaceutics Complete Blood Count Dataset provides structured, de-identified real-world data from over 600 US based laboratories, capturing diagnostic activity from CBC lab tests. Each record reflects a unique test event, including the date results were reported (TEST_DATE).

Test-level results are captured at the LOINC level with fields such as RESULT_NAME and RESULT_VALUE, alongside clinical context fields like RESULT_REFERENCE_RANGE and RESULT_UNITS_OF_MEASURE. Patient-level data includes a hashed patient ID (PATIENT_ID) and gender (GENDER). The physician associated with the test is captured through PHYSICIAN_NPI.

This dataset enables users to define and analyze cohorts based on test name, LOINC, result thresholds, associated physician, gender, and year of birth. Diaceutics has the ability to support linkage (via Datavant tokens) to diagnosis codes or therapy exposures supports advanced stratification for real-world evidence (RWE) and health economics (HEOR) use cases. Common applications include diagnostic access planning, trial feasibility analysis, and test-defined patient segmentation.

Data is delivered through the Snowflake Marketplace under a contact-based access model. Diaceutics provides the dataset directly to your Snowflake environment with the ability to support deliveries as frequently as weekly.

The CBC panel provides key hematology measures used to assess general health, monitor disease progression, and support clinical decision-making. Results include:

- White Blood Cell (WBC) Count
- Red Blood Cell (RBC) Count
- Hemoglobin & Hematocrit
- Mean Corpuscular Volume (MCV), Mean Corpuscular Hemoglobin (MCH), and Mean Corpuscular Hemoglobin Concentration (MCHC)
- Platelet Count and Differential (Neutrophils, Lymphocytes, Monocytes, Eosinophils, Basophils)

原始信息汇总

Diaceutics Complete Blood Count Dataset

数据集概述

Diaceutics Complete Blood Count Dataset 提供来自美国600多家实验室的结构化、去标识化真实世界数据，捕获CBC实验室测试的诊断活动。每个记录反映一个独特的测试事件，包括结果报告日期（TEST_DATE）。

数据时间范围

2016年1月1日至2025年9月17日

数据来源

覆盖美国所有州，来自600多家美国实验室

数据内容

测试级别数据

• LOINC级别的测试结果
• 结果名称（RESULT_NAME）和结果值（RESULT_VALUE）
• 临床背景字段：结果参考范围（RESULT_REFERENCE_RANGE）和结果测量单位（RESULT_UNITS_OF_MEASURE）

患者级别数据

• 哈希患者ID（PATIENT_ID）
• 性别（GENDER）
• 出生年份（YEAR_OF_BIRTH）

医生信息

• 医生NPI号码（PHYSICIAN_NPI）
• 医生专业（PHYSICIAN_SPECIALTY）

CBC面板包含的关键血液学指标

• 白细胞计数
• 红细胞计数
• 血红蛋白和血细胞比容
• 平均红细胞体积、平均红细胞血红蛋白、平均红细胞血红蛋白浓度
• 血小板计数和分类计数

业务应用场景

受众激活

基于实时实验室测试行为定位医疗专业人员

需求预测

按测试、生物标志物或地理位置跟踪真实世界诊断需求

生命科学商业化

通过基于诊断的洞察加速上市规划和市场执行

患者360视图

访问去标识化、实验室锚定的患者视图

市场分析

分析诊断市场份额、测试采用率和实验室覆盖率

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策划RWD

数据交付

通过Snowflake Marketplace基于联系访问模型交付，支持每周交付频率

数据字典

表名：MARKETPLACE_CBC_LAB_DATA

主要字段

• PATIENT_ID：Varchar
• GENDER：Varchar
• YEAR_OF_BIRTH：Number
• PHYSICIAN_NPI：Varchar
• PHYSICIAN_SPECIALTY：Varchar
• TEST_ID：Varchar
• TEST_DATE：Date
• LOINC_NAME：Varchar
• LOINC_CODE：Varchar
• TEST_NAME：Varchar
• RESULT_ID：Varchar
• RESULT_NAME：Varchar
• RESULT_VALUE：Varchar
• RESULT_REFERENCE_RANGE：Varchar
• RESULT_UNITS_OF_MEASURE：Varchar

使用示例

接收CBC测试的患者数量

sql SELECT COUNT(DISTINCT PATIENT_ID) AS PATIENT_COUNT FROM SNOWFLAKE.MARKETPLACE_CBC_LAB_DATA WHERE TEST_ID IS NOT NULL;

按专业统计医生数量

sql SELECT PHYSICIAN_SPECIALTY, COUNT(DISTINCT PHYSICIAN_NPI) AS PHYSICIAN_COUNT FROM SNOWFLAKE.MARKETPLACE_CBC_LAB_DATA WHERE PHYSICIAN_SPECIALTY IS NOT NULL GROUP BY 1 ORDER BY PHYSICIAN_COUNT DESC;

执行CBC测试的医生数量

sql SELECT COUNT(DISTINCT PHYSICIAN_NPI) AS PHYSICIAN_COUNT FROM SNOWFLAKE.MARKETPLACE_CBC_LAB_DATA WHERE TEST_ID IS NOT NULL;

每位患者的平均测试事件数

sql SELECT AVG(TEST_COUNT) AS AVERAGE_TESTS_PER_PATIENT FROM ( SELECT PATIENT_ID, COUNT(TEST_ID) AS TEST_COUNT FROM SNOWFLAKE.MARKETPLACE_CBC_LAB_DATA GROUP BY PATIENT_ID );

血小板计数临床显著范围的患者

sql SELECT * FROM SNOWFLAKE.MARKETPLACE_CBC_LAB_DATA WHERE TEST_ID IS NOT NULL AND RESULT_NAME = PLATELET COUNT AND (TRY_CAST(RESULT_VALUE AS FLOAT) >= 450 AND TRY_CAST(RESULT_VALUE AS FLOAT) <= 150);

技术支持

销售联系：[e-mail] 支持联系：[e-mail]

数据更新

静态数据

云区域可用性

AWS多个区域可用，包括：

• Asia Pacific (Mumbai)
• Asia Pacific (Osaka)
• Asia Pacific (Seoul)
• Asia Pacific (Singapore)