Frequently Asked Questions

1.Which tables are best for viewing all patient encounters in Best Practice (BP) and Medical Director (MD)?

For BP: The "visit" table is the best initial table for viewing all GP-patient encounters. If you want to specifically see GP visits based on the reason for the visit, then the "visit reasons" table would be more appropriate.
For MD: The tables to view patient encounters are CONSULTATION, DIAGNOSIS, PROGRESS, HISTORY, I_PRESCRIPTION, and VISIT. Depending on the condition, a combination of these tables may be necessary.

Refer to the Data Dictionary for more details.

2. Is identifying cohorts using Electronic Health Records (EHRs) a complex task?

Yes, identifying cohorts using EHR data, also known as electronic phenotyping in digital health research, can be challenging. This process involves defining specific criteria within EHRs to create patient groups for research purposes. For further information on Electronic Phenotyping, please refer to the provided article here.

3. Where can I learn more about electronic phenotyping?

The provided article here offers insights into electronic phenotyping and the challenges it presents within digital health research.

4. Does the Best Practice Past History table only include information collected during the first visit?

No, the PastHistory table may contain information documented during the initial visit or added during subsequent visits. This table captures previously diagnosed or observed conditions, procedures, or findings.

5. How can I verify the timing of information in the Past History table?

To validate the timing of information, we recommend cross-referencing the earliest visit date for each patient (found in the VISIT table) with the YEAR, MONTH, and DAY columns within the PastHistory table. This will provide insights into when the historical information was recorded.

6. How can I extract demographic data for patients aged 40 and over in Best Practice (BP) and Medical Director (MD)?

To simplify this process, the Patient table is recommended for use with both BP and MD data. After joining the Patient table with the respective Visit or Consultation table, calculate age using the DOB (date of birth) and then filter for patients 40 years of age or older. Finally, join this filtered data with the demographic data available in the Patients table.

7. Which tables should I consider when analyzing patients with hearing loss-related visits?

Best Practice (BP): We recommend examining the VISIT_REASON table for visit-related information. Additionally, consider including the PastHistory table for capturing previous diagnoses.
Medical Director (MD): Analyze data from the MD_PROGRESS and DIAGNOSIS_V1 tables for visit details. You may also want to include the HISTORY table for capturing past diagnoses. For even more specific patient identification, explore MBS/PBS codes.

8. Are analyses generally faster using OMOP? What are the pros and cons of OMOP?

Due to its standardized data formats and tools like ATLAS for cohort creation, analyses can be faster using OMOP. Some advantages of OMOP include standardized data, reduced errors, and the ability to conduct systematic analysis. However, there is a significant learning curve associated with OMOP, and harmonizing datasets to the CDM (Common Data Model) requires substantial resources. It's important to consider adopting OMOP as a separate project with its own dedicated funding and resources.

9. What does the "age at extraction" column in the Best Practice Patient's Table represent?

The "age at extraction" column signifies the patient's age at the time the data was extracted by the GRHANITE tool. To determine the patient's age at the time of visit, subtract the visit date from the patient's year of birth.

10. Can Best Practice and Medical Director datasets be considered independent for analysis?

While there may be some overlap in terms of patients, it is reasonable to treat the Best Practice and Medical Director datasets as independent for analysis purposes. This is because the datasets were extracted independently, and there can be errors and inconsistencies between them. To identify shared or duplicate patients and acknowledge potential discrepancies in your analysis, it's recommend to perform data linkage using the GRHANITE linker tool.

11. Is ethnicity data available for patients in the Medical Director Module?

Unfortunately, ethnicity data in general practice datasets, including Medical Director, is often limited. The captured data may include categories like "Blank," "Not Stated," or "Inadequately Described."

12. Is ethics approval required to access the data? And if so, what is the typical timeframe for approval?

Yes, obtaining ethics approval is mandatory before accessing the data. The approval process typically takes up to 2-3 weeks, although this timeframe may vary depending on the specific project details. For further information on the ethics approval process, please refer to the SREDH Consortium Governance documents.