Question (as you put it to us, in our words): you have lows more often
after highs, and the lows bring the stronger symptoms. If highs lead to lows,
would keeping the highs down prevent them?
Interpretation of the Question (an attempt to
make the question slightly more precise, computable and scientific)
“A high” we are interpreting as a rise that peaked above
180 mg/dL, the top of the target range doctors use for people with
diabetes1. Because you are not on insulin, we also checked a lower
line, 140, which healthy adults stay under almost all the time2.
“A low” we are interpreting as below 70 mg/dL for at least 15
minutes, ending once you are back above 70 for 15 minutes3.
“After” we are interpreting as a low starting between 1 and 5 hours
after the high peaked.
Two readings of your question, both decided before any result was seen: are
lows more likely after a high, and do bigger rises bring
deeper dips?
Answer to the Presenting Question: No
No, based on the analysis below, and within
its assumptions and limits. We are fairly confident of that.
Only 1 of your 39 lows came within 5 hours
of a reading above 180. So even if every high caused a low, keeping your highs
under 180 would have prevented at most 1.
Most of your lows (24 of 39) started between midnight and 6 AM,
hours after any meal.
Bigger rises did not bring deeper dips afterwards.
If your highs brought on your lows, a high in the 5 hours before a low would
be commoner than a high in the 5 hours before an ordinary moment. That is not what your
readings show.
Counting a high as above 180: 3%
of your lows had one in the 5 hours before. At ordinary moments,
2% did.
Counting a high as above 140: 23%
of your lows had one in the 5 hours before. At ordinary moments,
32% did.
“An ordinary moment” is the same time of day on each of your
other days in the same stretch of wear, so the same sensor, the same hour and the same
daily rhythm. Hover a bar for its count.
When do your lows start?
Mostly at night. 24 of your 39 lows started between midnight
and 6 AM, hours after any meal. Your highs come after meals, in the afternoon and evening.
Each dot is one of your 39 lows, at the hour it started. A filled dot
means there was a reading above 140 in the 5 hours before; a hollow dot means there
was not. The shaded band is midnight to 6 AM. Hover a dot for its hour.
Data Source(s) and Date(s)
Data sources:
Devices:FreeStyle LibreFreeStyle Libre is a line of CGM sensors made by Abbott; the number after the name is the model. More at freestyle.abbottCGMA CGM (continuous glucose monitor) is a small sensor worn on the arm that reads glucose every few minutes, day and night, for about two weeks. More at en.wikipedia.org/wiki/Continuous_glucose_monitordata, exported
September 2026, read through the FreeStyle LibreFreeStyle Libre is a line of CGM sensors made by Abbott; the number after the name is the model. More at freestyle.abbott 3 and then the Libre appThe Libre app is Abbott's phone app that reads a FreeStyle Libre sensor and uploads its readings. In an export it names the software that uploaded a reading, not the sensor that took it. More at freestyle.abbott. The older FreeStyle LibreFreeStyle Libre is a line of CGM sensors made by Abbott; the number after the name is the model. More at freestyle.abbott 2 readings are
left out, as on your other pages, and so is each sensor's first day, which reads low.
Textual sources: your question, as you told it to us (October 2026), and
that you are not on insulin (October 2026).
Dates: 121 days of readings, February 2023 to July 2026, in
8 stretches of wear.
People and Organizations Named or Anonymized:
Named (in your template names, your notes, or what you told us): None
Anonymized (shown by alias, inside your own notes and quotes too): None
What we compared (independent variable)
Every time your glucose rose at least 30 mg/dL
to a peak (353 rises), split two ways:
rises that peaked above 180 (17)
rises that peaked at 180 or below (336): meals that did not
spike as high, which is the change you would be making
What we measured (dependent variable)
whether a low started 1 to 5 hours after the peak
how far you dipped below where the rise started, in those same hours
Lows after a high: No
After a rise above 180, a low followed 1 time
in 17. After the other rises, 10 times
in 336. That difference is well within what chance makes (statistical
significance: no).
How much this test could see: you had only 17 rises above
180. With so few, only a very large effect, about 10 times the risk, would
reliably show up. So this test alone cannot rule out a smaller effect, which is why the
answer rests mostly on the look-back above. With the lower line, 140, there
are 181 high rises, and lows followed
6/181 of them against
5/172 of the others: about the same.
Bigger rises, deeper dips: No
No. Across your 353 rises, the bigger ones were not followed by
deeper dips; if anything slightly the reverse (rank correlation -0.10). This
test had enough rises to see an effect of the size we said would matter.
Day by day, the same holds: on days with a higher peak, your lowest
reading was a little higher, not lower (rank correlation
+0.15, 99
full days).
Caveats
These are your sensor's lows. Lows at night on a CGMA CGM (continuous glucose monitor) is a small sensor worn on the arm that reads glucose every few minutes, day and night, for about two weeks. More at en.wikipedia.org/wiki/Continuous_glucose_monitor are sometimes false,
from lying on the sensor. 24 of your 39 started between midnight and
6 AM. Leaving those out does not change the answer:
40% of your daytime lows came after a reading
above 140, against 41% of
ordinary moments.
We cannot see which lows you felt. If your symptoms come with only some of
them, the answer for those could differ. See the question at the end.
Your highs and lows mostly come from different sensors. 16 of your 17 rises above
180 were read through the FreeStyle LibreFreeStyle Libre is a line of CGM sensors made by Abbott; the number after the name is the model. More at freestyle.abbott 3; most of your lows were not. Each
comparison was made within the same device and time of day, so this cannot make the
result, but it does leave fewer like-for-like rises.
The 180 and 70 lines were set for people with diabetes.1
For someone not on insulin, doctors only call it hypoglycaemia below about
55 with symptoms4. At 55, no low followed any of your
rises above 180.
This describes a pattern in your readings. It is not medical advice.
One thing worth a closer look
After your rises above 180, you more often dipped at least 20 mg/dL
below where the rise had started, without going below 70:
4/17 times, against 28/336 after the other rises.
This is a lead, not a finding. It was one of many extra checks, it does
not pass the bar for chance once those are counted, and it rests on 17 rises. If your
symptoms come with drops like these rather than with true lows, this is the version of
the question to test next, on new readings.
References
Battelino T, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the International Consensus on Time in Range. Diabetes Care 2019;42:1593-1603. https://doi.org/10.2337/dci19-0028(read in full; its targets are stated for people with type 1 and type 2 diabetes)
Shah VN, et al. Continuous glucose monitoring profiles in healthy nondiabetic participants: a multicenter prospective study. J Clin Endocrinol Metab 2019;104:4356-4364. https://doi.org/10.1210/jc.2018-02763(read in full)
Danne T, et al. International Consensus on Use of Continuous Glucose Monitoring. Diabetes Care 2017;40:1631-1640. https://doi.org/10.2337/dc17-1600(read in full; section 4 defines when a low begins and ends)
Cryer PE, et al. Evaluation and management of adult hypoglycemic disorders: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2009;94:709-728. https://doi.org/10.1210/jc.2008-1410(the 55 mg/dL figure as summarised by others; the full text was not read)
Both questions were written down, with every definition, before any of
this was counted. Each was tested by shuffling which rises were “high” 10,000
times, only among rises from the same device and the same 3 hours of the day, and
counting how often chance made a result as strong. Corrected for two questions (Holm).
Question
Result
p
corrected p
Verdict
Lows after a high
1/17 vs
10/336; risk ratio 2.0;
detectable: 10x
0.40
0.80
NULL
Bigger rise, deeper dip
rho -0.10; detectable:
0.15
0.93
0.93
NULL
Extra checks (leads only, 2,000 shuffles each):
Check
Low after high
after other
p
rho
high 140
6/181
5/172
0.34
-0.10
low 55
0/17
3/340
1.00
-0.09
high 140, low 55
2/182
1/175
0.44
-0.09
night lows (00-06) removed
1/17
6/340
0.28
-0.10
night lows removed, high 140
4/181
3/176
0.42
-0.10
compression-shaped lows removed
1/17
9/336
0.38
-0.10
warm-up 0 h
1/19
11/373
0.39
-0.10
device: FreeStyle LibreFreeStyle Libre is a line of CGM sensors made by Abbott; the number after the name is the model. More at freestyle.abbott 3
1/16
4/216
0.34
-0.15
device: Libre appThe Libre app is Abbott's phone app that reads a FreeStyle Libre sensor and uploads its readings. In an export it names the software that uploaded a reading, not the sensor that took it. More at freestyle.abbott
0/1
6/120
1.00
+0.02
relative low (>= 20 below the pre-rise low)
4/17
28/336
0.06
-0.10
What do you want to know next?
Each one runs on your own export. What you asked is written down with today's date before anything is computed, so the question stays separate from the answer.
Which of your lows did you feel?
This page counts the lows your sensor saw. If you tell us roughly when you felt low, we can ask the same question of just those.
Rough dates and times are enough. Only we see them; they are not put on the page.
Ready when you tell us
If you have a health-related question about your own data,join the waitlist for your own custom analysis.
Act soon: it is free while we stand this service up and refine it. All we ask in return is permission to publish a results page like the ones we have made for our other clients. We never publish your surname or your date of birth. The page shows a first name or a pseudonym, whichever you prefer, and your age in years and your sex.
Made by PerSci from your CGMA CGM (continuous glucose monitor) is a small sensor worn on the arm that reads glucose every few minutes, day and night, for about two weeks. More at en.wikipedia.org/wiki/Continuous_glucose_monitor export. It describes a pattern in your readings; it is not medical advice.
PerSci is not medical advice. Analyses describe one person's own data and are not a diagnosis or a treatment recommendation. Full disclaimer · Privacy Policy