You can export blood-pressure data already stored in Apple Health to CSV or JSON with Health.md on iPhone. Choose daily statistics for a summary, or enable Detailed Time-Series for available timestamped systolic/diastolic pairs. Manual export needs no Mac.
Choose CSV, JSON, or a PDF
| What you need | Start here |
|---|---|
| Sort or chart exported values | CSV, following the steps below. |
| Process paired readings with separate numeric fields | JSON, with Detailed Time-Series enabled. |
| Share a PDF of a completed Apple Blood Pressure Log | Apple's built-in Export PDF option. See Apple's log instructions for requirements and availability. |
| Create a selected, multi-measurement PDF in Health.md | Clinician Report walkthrough. |
Apple's log PDF and Health.md's data exports are different workflows. This guide covers files for your own inspection or processing; it does not interpret readings or recommend treatment. For Apple's full XML archive, see the general Apple Health export guide.
1. Select metrics and preview one day
- Confirm the readings you want are present in Apple Health on this iPhone. Let your source app finish saving or syncing them first.
- In Health.md, authorize the relevant health reads. Under Health Metrics → Vitals, select Blood Pressure (Systolic) and Blood Pressure (Diastolic).
- Choose CSV or JSON. Use Summary for daily statistics; enable Detailed Time-Series when you also need individual paired readings.
- Choose one known day, use YYYY-MM-DD as the export date format for easy checking, and select a writable Files folder.
- Tap Preview, check the dates, units and intended rows, then export. Open the saved file to confirm its contents and destination.
A longer range creates one file per day. Use the first iPhone export guide for permissions and folder setup, or troubleshooting if Preview succeeds but saving fails.
Daily systolic and diastolic statistics come from separate quantity queries. Detailed readings come from actual HealthKit blood-pressure correlations, which group the two components. Health.md does not construct pairs by joining unrelated systolic and diastolic values. Apple's blood-pressure correlation reference describes this paired representation.
2. Read the CSV without mixing statistics and readings
Inspect the fictional daily CSV example. Its header is:
Date,Category,Metric,Value,Unit,Timestamp
These exact rows show two daily statistics and one paired reading:
2026-03-15,Vitals,Blood Pressure Systolic Avg,121.0,mmHg
2026-03-15,Vitals,Blood Pressure Diastolic Avg,79.0,mmHg
2026-03-15,Vitals,Blood Pressure Sample,121/79,mmHg,2026-03-15T01:23:20Z
Filter Category = Vitals and Unit = mmHg, then choose the exact Metric:
- Blood Pressure Systolic Avg or Blood Pressure Diastolic Avg for a daily average series. Separate Min and Max metrics may also be present.
- Blood Pressure Sample for individual paired readings. Its Value is text in systolic/diastolic order, such as
121/79.
Keep the paired Value column as text when importing into a spreadsheet so a value containing / is not interpreted as a date. If you split it into two numeric columns, label them systolic and diastolic and retain the Timestamp. Do not turn 121/79 into a fraction or average it as one number.
The full Value column is mixed-type. Filter the intended metrics before converting cells to numbers. Summary rows above omit the final Timestamp field; the sample row has all six fields. Use a CSV-aware importer that preserves quoted fields and leaves an omitted final cell empty.
For spreadsheet import steps, use Excel or Google Sheets, applying the blood-pressure filters above. Keep daily averages and individual readings in separate series.
3. Use JSON for separate numeric components
The fictional daily JSON example contains a nested vitals object. Daily fields include bloodPressureSystolicAvg, bloodPressureDiastolicAvg, and their Min/Max counterparts when available. The compatibility fields bloodPressureSystolic and bloodPressureDiastolic repeat the daily averages; they are not additional readings.
With detailed data available, vitals.bloodPressureSamples contains objects with separate numeric systolic and diastolic fields, timestamp, endDate, unit, and available identity/source metadata. In the fictional example, one object has systolic 121, diastolic 79, and unit mmHg. Those values demonstrate file structure, not a clinical example or a new device export.
Inspect the actual output's units and time_context. A timestamp ending in Z is UTC; the daily Date label follows the export calendar. Keep timezone conversion explicit when displaying readings. An omitted field or empty array is not a measured zero.
The example tests serializer output, not a complete day from which every daily statistic can be recomputed. Separate summary queries and paired-record queries can have different coverage. Do not assume that combining the two daily averages recreates a recorded pair.
Check missing readings and saved copies
- Statistics but no paired rows: check Detailed Time-Series, selected metrics, the source day and permissions. A source may contain separate quantities without the correlations needed for paired rows.
- No data: confirm records, synchronization, date range and read access. An empty result does not prove that no measurement occurred or reveal whether access was denied.
- Unexpected dates: compare the export calendar and UTC timestamps before moving records between days.
- Duplicate files: keep the intended version of a re-exported day. Do not count repeated exports as new readings.
The local folder workflow requires no Health.md account and creates no Health.md server copy by default. Synced folders and recipients govern their own copies. Inspect detailed timestamps and source information before sharing.
The separate opt-in, single-owner Health.md Cloud pilot retains only API exports intentionally uploaded to it, has no public signup and no automatic device sync, and is not generally available. This folder workflow does not upload to that pilot. See retention and destination boundaries.
For a non-synced destination, follow export health data without the cloud. Android users should start with the Health Connect export guide; this article's Apple field names and query behavior are not an Android contract.