Clinical Lab Tests

Overview of Data Domain

Clinical lab tests are a fundamental component of medical diagnostics and patient care. They encompass a wide range of tests performed on samples of blood, urine, tissue, or other bodily substances to assess health and diagnose disease. The data domain for clinical lab tests includes several key elements:

1. Test Name and Description: Each lab test has a specific name and purpose, which provides vital information about what aspect of health or disease is being evaluated.

2. Interpretation: The analysis and understanding of test results in the context of a patient’s overall health and medical history. Abnormal results may indicate an underlying condition that requires further investigation or treatment.

3. Measurement Units: Lab tests results are reported in specific units of measure, such as milligrams per deciliter (mg/dL) for blood sugar levels or cells per microliter for blood cell counts.

4. Reference Ranges: These are established norms used to compare individual test results. They indicate what is considered normal for a specific population or individual, based on age, sex, and other factors.The reference ranges for all clinical lab tests performed by the UW NORC lab (and CBC test performed at local labs for each Data site).

5. Method of Collection This refers to how the sample is obtained, whether it's a blood draw, urine collection, or tissue biopsy.

Variables included in Data Domain

See the table here

NT-proBNP Reference Intervals

Sex	Age_Low	Age_High	Range_Low	Range_High
F	0d	2d	321	11987
F	3d	10d	263	5918
F	11d	1m	263	5918
F	2m	1y	37	646
F	2y	2y	39	413
F	3y	6y	23	289
F	7y	14y	0	157
F	15y	18y	0	158
F	19y	38y	0	160
F	40y	44y	0	162
F	45y	54y	0	141
F	55y	64y	0	226
F	65y	200y	0	540
M	0d	2d	321	11987
M	3d	10d	263	5918
M	11d	1m	263	5918
M	2m	1y	37	646
M	2y	2y	39	413
M	3y	6y	23	289
M	7y	14y	0	157
M	15y	18y	0	158
M	19y	38y	0	79
M	40y	44y	0	72
M	45y	54y	0	87
M	55y	64y	0	88
M	65y	200y	0	540

Alkaline Phosphatase reference ranges

Sex	Age_Low	Age_High	Range_Low	Range_High
F	0y	9y	111	281
F	10y	11y	132	366
F	12y	13y	89	285
F	14y	17y	43	226
F	18y	24y	26	98
F	25y	34y	25	100
F	35y	44y	25	112
F	45y	54y	34	121
F	55y	64y	31	132
F	65y	74y	38	172
F	75y	200y	49	199
M	0y	9y	115	324
M	10y	11y	115	324
M	12y	13y	119	426
M	14y	17y	72	400
M	18y	24y	42	136
M	25y	34y	35	109
M	35y	44y	36	122
M	45y	54y	39	139
M	55y	64y	37	159
M	65y	74y	36	161
M	75y	200y	52	227

Data Processing

File Format

Data was exported from the RedCap Survey that was filled by clinical personnel into .csv format.

File organization is as follows:

pilot_data_root
└── clinical_data
   └── measurement.csv

Domain	Variable	Data Capture	Method	Data Standard/File Extension	Open Source vs. Protected Database?
Labs	measurement.csv	device, then Azure	RedCap	.csv	Open source

Data Standards

OMOP

The OMOP (Observational Medical Outcomes Partnership) data standard, developed by the Observational Health Data Sciences and Informatics (OHDSI) program, is a standardized framework designed to improve the quality, reliability, and comparability of observational healthcare data. The main goal of the OMOP standard is to enable the aggregation and analysis of healthcare data from diverse sources, such as electronic health records (EHRs), insurance claims, and registries, in a consistent and interoperable manner.

Reference : https://ohdsi.github.io/CommonDataModel/

File Processing

The .csv file can be opened and processed using notebook application or any coding language example:( Python, R). Here is a snippet code to extract HbA1c.

import pandas as pd

# Load the CSV file
git reset head# Replace 'your_file.csv' with the path to your CSV file

# Replace 'HBA1C_IDENTIFIER' with the actual identifier for HbA1c in your dataset
HBA1C_IDENTIFIER = 'HbA1c (%)'  # This should be the actual identifier used in your CSV

# Filter the dataframe for rows where measurement_source_concept_id is HbA1c
hba1c_df = df[df['measurement_source_value'] == HBA1C_IDENTIFIER]

# Selecting only the necessary columns
hba1c_df = hba1c_df[['person_id', 'value_as_number', 'measurement_source_value']]

# Save the filtered data to a new CSV file
hba1c_df.to_csv('hba1c.csv', index=False)

print(‘ Data saved to hba1c.csv')

Metadata and Example Outputs

Variable	Description	Example 1	Example 2
measurement_id	Unique identifier for the measurement	10769	10770
person_id	Unique identifier for the person	1099	1500
measurement_concept_id	Concept ID for the type of measurement	3035995	3004410
measurement_date	Date of the measurement	2023-11-30	2023-07-27
measurement_datetime	Date and time of the measurement	2023-11-30 00:00:00	2023-07-27 00:00:00
measurement_time	Time of the measurement	00:00:00	00:00:00
measurement_type_concept_id	Concept ID for the type of measurement	32856	32856
operator_concept_id	Concept ID for the operator of measurement	4172703	4172703
value_as_number	Numeric value of the measurement	65.0	5.7
value_as_concept_id	Concept ID for the value of measurement	0	0
unit_concept_id	Concept ID for the unit of measurement	0	0
range_low	Lower range of the measurement value	34.0	4.0
range_high	Higher range of the measurement value	139.0	6.0
provider_id	Unique identifier for the provider	0	0
visit_occurrence_id	Unique identifier for the visit occurrence	0	0
visit_detail_id	Unique identifier for the visit detail	0	0
measurement_source_value	Source value of the measurement	Alkaline Phosphatase (IU/L)	HbA1c (%)
measurement_source_concept_id	Concept ID for the measurement source value	0	0
unit_source_value	Source unit of the measurement	IU/L	%
unit_source_concept_id	Concept ID for the unit source	0	0
value_source_value	Source value for the measurement value	65	5.7
measurement_event_id	Unique identifier for the measurement event	0	0
meas_event_field_concept_id	Concept ID for the measurement event field	0	0