Picture a kitchen table on Sunday night. The monitor shows a rolling average. A paper log has one missed evening. Several extra readings were taken after a surprising screen. A family member wants one clean number for Monday's appointment. If everything gets fed into a calculator without a plan, the result may look tidy while the record underneath it becomes harder to interpret.
A useful average should shorten the story, not rewrite it.
First decide what job the average is doing
Home blood pressure is not collected for one universal reason. It may help a clinician compare home and office patterns, review a treatment plan, or understand readings around symptoms. The U.S. Preventive Services Task Force recommends out-of-office measurements to confirm a hypertension diagnosis before treatment begins. That is a clinical use of a structured record, not permission to diagnose yourself from a homemade average.
Before pressing any calculator button, write one sentence at the top of the log:
If a clinician supplied the dates, timing, arm, medication relationship, number of readings, or first-day rule, copy those instructions exactly. If no plan was supplied, ask what the office wants before building a diagnostic series. Our guide to how often to check blood pressure at home explains why a short baseline schedule and long-term maintenance schedule are different jobs.
Three summaries are often confused:
- A session average summarizes the planned readings taken at one sitting.
- A monitoring-period average summarizes included readings across a fixed group of days.
- A monitor's memory average uses whatever profile, date window, and stored entries that device is programmed to include.
Six steps that preserve both the summary and the evidence behind it.
Trend review, clinician-requested series, or a same-day concern are not interchangeable.
Write dates, timing, planned repeats, and any directed exclusions before seeing the results.
Preserve every result, blank session, device error, and useful context note.
Use the prewritten rule. Do not keep only the readings you hoped to see.
Average systolic and diastolic separately over the same included rows.
Share the average, dates, count, schedule, notes, and raw readings together.
Name the purpose. Lock the collection rule before seeing results. Keep the raw log. Sort readings without cherry-picking. Average systolic and diastolic separately. Report the result with dates, count, schedule, notes, and raw readings.
Source note: This TrueVitals USA process synthesizes the American Heart Association's paired-reading and recording guidance, the AAFP multi-day home-monitoring review, and the AMA's separate systolic/diastolic averaging tool. Protocol-specific first-day handling is discussed below.
How to calculate a home blood pressure average
1. Mark the included rows
Work from the original log or device memory, not from numbers recopied from memory. Mark each row that belongs under the rule chosen at the start. A missing session stays missing. Never enter zero, copy a neighboring reading, or take extra measurements the next day to make the grid look complete.
2. Count complete reading pairs
Each included row should contain one systolic value and its matching diastolic value from the same measurement. Call the number of included rows n. If a transcription is missing half of a pair, check the device memory or original record. Do not match a top number from one reading with a bottom number from another.
3. Add and divide each column separately
Do not average systolic and diastolic together. They remain a pair in the final summary, such as “average SYS / average DIA.” Do not give a daily average and a single raw reading equal weight either. One may represent several measurements while the other represents one.
4. Keep the calculation auditable
Write down the two sums, the included-reading count, and the result before rounding. If a form wants whole numbers, keep the more precise calculation in your notes and round only the final display. The important safeguard is that someone else can reproduce the result from the same marked rows.
Use one row for each included measurement. Add more rows as needed.
| Included row | Systolic | Diastolic | Context, if needed |
|---|---|---|---|
| 1 | |||
| 2 | |||
| … | |||
| n | |||
| Sum | n = | ||
| Average | sum ÷ n = | sum ÷ n = |
Source note: The worksheet follows the same separate-column arithmetic used by the AMA BP Average Calculator. It intentionally contains no invented patient readings or treatment targets.
Which readings belong in the average?
This is where quiet cherry-picking can creep in. “Valid” does not mean “close to what I expected.” It means the measurement was obtained under the chosen protocol without a known failure that the protocol says to exclude.
| What happened | What to preserve | What to do with the average |
|---|---|---|
| Planned reading with clean setup | The full systolic/diastolic pair and time | Include it under the written plan. |
| Missed session | A blank marked “missed” | Do not enter zero or manufacture a replacement. |
| Monitor error with no result | The error and what was corrected | There is no blood pressure pair to calculate. |
| Talking, movement, slipped cuff, or another known disruption | The displayed result, if any, plus the disruption note | Apply the prewritten validity rule or ask the care team. Do not silently erase or include it. |
| Valid result that is simply higher or lower than expected | The result and relevant context | Do not discard it solely because of its direction. |
| First day of a multi-day series | Every day-one result | Include or exclude it only according to the selected protocol. Excluded does not mean deleted. |
| Extra repeats taken to look for a lower number | All results in order | Do not improvise a favorable subset. Ask which planned sequence should be summarized. |
| Symptoms or a threshold in a written action plan | The reading, symptoms, time, and action taken | Follow the action plan. Do not wait for a weekly average. |
Evidence note: The AHA advises taking two readings one minute apart and recording the results. The NICE diagnostic HBPM protocol excludes the first day, while the 2021 European Society of Hypertension position paper describes first-day exclusion as debated and says its impact appears smaller with more monitoring days. That is why the protocol must be named rather than guessed. For a deeper distinction between a first reading, first day, and failed first attempt, see our first-reading evidence guide.
What an average can hide
One combined number can conceal a consistent morning-evening difference, a run of readings taken before versus after medication, a gradual change across days, or a result connected to symptoms. None of that makes averaging useless. It means the raw sequence still matters.
Keep morning and evening labels. Keep medication timing only if your clinician asked for it or it helps explain the collection plan. Keep notes short and factual: “missed evening,” “talked during attempt,” “new dizziness,” or “different arm by clinic instruction.”
Give the clinician an average plus an evidence packet
The most useful handoff is not a lone pair of numbers. It is a compact summary that lets the office understand how that pair was produced.
TrueVitals customers can use the TrueVitals BP Log at no charge to keep readings and short context notes together. Whether you use an app or paper, bring the monitor itself when a clinician wants to compare its memory with the record.
Edge cases that change the answer
Your monitor calculates an average for you
Check the manual before trusting the label. Confirm the user bank, date window, number of stored readings, and whether guest or older entries are included. A rolling device average may not match the fixed date range your clinician requested. Our guide to blood pressure monitor memory shows what to inspect.
Two people share one monitor
Use separate user profiles or unmistakably separate logs. A mathematically correct household average is clinically useless if it blends two people. If the device has only one memory bank, record outside the device immediately and verify names before the appointment.
Only one reading in a planned pair worked
Keep that valid reading and mark the missing partner. Do not copy the first result into the second slot. Whether the single reading belongs in the final calculation depends on the protocol and why the series was ordered.
The average looks fine, but one result came with symptoms
Report that episode separately. A weekly mean can be arithmetically correct and still be the wrong tool for a same-day concern. Use the clinician's action plan or urgent-care instructions rather than waiting for more data to pull the average in either direction.
The paper log and device memory disagree
Do not quietly “fix” one source. Mark the mismatch, bring both, and look for a wrong date, wrong user bank, transcription error, or extra stored attempt. The disagreement is useful troubleshooting evidence.
Frequently asked questions
Do I average systolic and diastolic blood pressure together?
No. Add the included systolic values and divide by the included-reading count. Then repeat the calculation separately for the matching diastolic values. Report the two averages as a pair.
Should I delete the highest home blood pressure reading before averaging?
No, not simply because it is the highest. Keep every raw result. Exclude a reading only under a preselected protocol or a documented validity rule, and retain the excluded result and reason in the record.
Should the first day of home blood pressure readings be averaged?
It depends on the protocol. Some diagnostic protocols exclude day one, while other guidance averages all measurements and notes that first-day exclusion is debated. Follow the rule supplied by your clinician and keep the day-one readings either way.
What should I do if I miss a blood pressure reading?
Leave the space blank, label it missed, and continue at the next planned session. Do not enter zero, copy another result, or take extra readings later just to fill the grid.
Can I use the average shown in my blood pressure monitor's memory?
Only after checking what it includes. Confirm the user profile, date window, stored-reading count, and whether old, guest, or extra readings are part of the device's calculation. Share the raw memory or export with the average.
References
- American Heart Association. Home Blood Pressure Monitoring. Last reviewed August 14, 2025.
- American Medical Association. BP Average Calculator.
- American Medical Association. What doctors want patients to know about home BP measurement. Updated March 13, 2026.
- Weinfeld JM, Hart KM, Vargas JD. Home Blood Pressure Monitoring. American Family Physician. 2021;104(3):237-243.
- National Institute for Health and Care Excellence. Hypertension in adults: diagnosis and management, recommendations.
- Parati G, Stergiou GS, Bilo G, et al. Home blood pressure monitoring: methodology, clinical relevance and practical application. Journal of Hypertension. 2021;39:1742-1767. doi:10.1097/HJH.0000000000002922.
- U.S. Preventive Services Task Force. Hypertension in Adults: Screening.
When the equipment makes the record harder than it should be
If cuff fit, screen readability, user memory, or repeatable setup is getting in the way, review the practical features in the TrueVitals Pro 365 specifications.