Sleep & CPAP Report — 2026-09-13
🚩 Flags
- Sharp architectural reversal — yesterday's two-week best has been fully undone in one night. Deep% 8.29% (vs 18.06% last night, -9.77 pp overnight) — the LOWEST deep% since 2026-09-07 (8.05) and second-lowest of the recent 10-night window. REM% 20.78% (vs 27.63%, -6.85 pp). Core% 70.93% (up +16.62 pp from 54.31% last night, HIGHEST core% of recent 10-night window). Sleep tonight was heavily dominated by light stages — the single-night reversal noted in the prior report as needing "one more good night" instead broke the wrong way.
- Awake time DOUBLED vs baseline — 18.06 min tonight, HIGHEST of recent 10-night window (prior range 2.5–13.5). +98.4% vs 7d avg (9.10) and +84.6% vs 30d avg (9.78). Combined with the collapsed deep/REM, this is a clearly fragmented night despite the seemingly-high 95.03% efficiency figure.
- AHI 2.60 — elevated vs baseline (+36.4% vs 7d avg 1.91, +33.5% vs 30d avg 1.95). Still well within clinically fine (<5), but this is the second-highest AHI of the recent 10-night window (only 09-04 at 3.37 was higher).
- OA count 14 — HIGHEST of recent 10-night window (prior range 7–13, next-highest was 09-11 at 13). +55.6% vs 7d avg (9.00), +47.9% vs 30d (9.47). The AHI rise is driven entirely by obstructive events, not centrals (CA=1) or hypopneas (H=1, actually the LOWEST H count of the recent window).
- Severe flow-limitation exposure spiked — Min_Above_020 = 7.6 min, +139.6% vs 7d avg (3.17), +85.2% vs 30d (4.10), and HIGHEST of recent 10-night window. FL_P95 0.11 also at recent-window high (tied with 09-05, 09-07, 09-09). Min_Above_010 19.1 min (+30.4% vs 7d, near recent-window high). The flow-limitation de-escalation flagged in prior reports has broken tonight — this is a distinct regime shift for FL exposure.
- Pressure response — machine pushed higher. Pressure_Avg 7.78 (+9.2% vs 7d, +3.8% vs 30d — HIGHEST since 09-04 at 7.37). Pressure_90 10.16 (+3.9% vs 7d). Consistent with the machine responding to the elevated obstructive/FL load.
- Leak seal REMAINS good — this is NOT a mask issue. Leak_Avg 0.47 (-25.9% vs 7d, -39.7% vs 30d). Leak_90 0.0 (perfect). Leak_Max 18 (normal band). Five consecutive nights with Leak_Avg well below the 30d baseline. The AHI/OA/FL spike tonight is happening in the ABSENCE of any leak deterioration — pointing to a physiological driver (position, congestion, alcohol, allergen, illness, muscle tone, etc.) rather than a hardware issue.
- Sleep shorter than baseline — TIB 363.3 min (-4.0% vs 7d, -5.9% vs 30d) and Total_Sleep 345.2 min (-6.5% vs 7d, -8.2% vs 30d). Also the shortest sleep of the recent 10-night window. Consistent with a compressed, low-quality night rather than a full run.
Summary Table — Night of 2026-09-13
| Metric |
Value |
| Sleep start |
2026-09-13 23:45:58-05:00 |
| Time in bed |
363.3 min (6h 3.3m) |
| Total sleep |
345.2 min (5h 45.2m) |
| Sleep efficiency |
95.03% |
| Awake time |
18.1 min |
| Core (N1+N2) sleep |
244.9 min / 70.93% |
| Deep (N3) sleep |
28.6 min / 8.29% |
| REM |
71.8 min / 20.78% |
| CPAP usage hours |
6.15 |
| Session count |
1 |
| AHI |
2.60 |
| RDI |
2.60 |
| Obstructive apneas (OA) |
14 |
| Hypopneas (H) |
1 |
| Central apneas (CA) |
1 |
| Unclassified apneas (UA) |
0 |
| RERA |
0 |
| Pressure_Avg |
7.78 |
| Pressure_Min |
4.00 |
| Pressure_Max |
12.32 |
| Pressure_90th |
10.16 |
| Leak_Avg |
0.47 |
| Leak_Max |
18.0 |
| Leak_90th |
0.00 |
| FL_Avg |
0.0170 |
| FL_P95 |
0.11 |
| FL_Max |
0.54 |
| Min_Above_005 |
38.3 |
| Min_Above_010 |
19.1 |
| Min_Above_020 |
7.6 |
| Frac_NonZero (FL samples) |
23.71% |
Comparison to Trailing Averages
| Metric |
Tonight |
7d avg |
30d avg |
Δ vs 7d |
Δ vs 30d |
| AHI |
2.60 |
1.91 |
1.95 |
+36.4% |
+33.5% |
| OA count |
14.00 |
9.00 |
9.47 |
+55.6% |
+47.9% |
| H count |
1.00 |
2.14 |
2.57 |
-53.3% |
-61.0% |
| CA count |
1.00 |
1.00 |
0.77 |
+0.0% |
+30.4% |
| Pressure_Avg |
7.78 |
7.13 |
7.50 |
+9.2% |
+3.8% |
| Pressure_90 |
10.16 |
9.78 |
10.07 |
+3.9% |
+0.9% |
| Leak_Avg |
0.47 |
0.63 |
0.78 |
-25.9% |
-39.7% |
| Leak_Max |
18.00 |
16.11 |
17.40 |
+11.7% |
+3.4% |
| Leak_90th |
0.00 |
1.37 |
2.08 |
-100.0% |
-100.0% |
| Usage hours |
6.15 |
6.39 |
6.55 |
-3.7% |
-6.1% |
| FL_P95 |
0.11 |
0.084 |
0.099 |
+30.5% |
+11.1% |
| Min_Above_010 |
19.10 |
14.64 |
18.37 |
+30.4% |
+4.0% |
| Min_Above_020 |
7.60 |
3.17 |
4.10 |
+139.6% |
+85.2% |
| Awake min |
18.06 |
9.10 |
9.78 |
+98.4% |
+84.6% |
| TIB min |
363.28 |
378.41 |
386.02 |
-4.0% |
-5.9% |
| Total sleep min |
345.22 |
369.31 |
376.23 |
-6.5% |
-8.2% |
| Deep % |
8.29 |
12.87 |
13.67 |
-35.6% |
-39.4% |
| REM % |
20.78 |
22.85 |
24.04 |
-9.0% |
-13.5% |
| Core % |
70.93 |
64.28 |
62.29 |
+10.3% |
+13.9% |
Week-over-Week Deltas (recent 7d 09/07–09/13 vs prior 7d 08/31–09/06)
| Metric |
Recent 7d |
Prior 7d |
Δ |
Δ % |
| AHI |
2.079 |
2.253 |
-0.174 |
-7.7% |
| OA |
10.286 |
11.286 |
-1.000 |
-8.9% |
| CA |
1.000 |
0.429 |
+0.571 |
+133.3% |
| Pressure_Avg |
7.284 |
7.503 |
-0.219 |
-2.9% |
| Leak_Avg |
0.627 |
0.902 |
-0.275 |
-30.5% |
| Leak_90 |
1.371 |
2.229 |
-0.857 |
-38.5% |
| Deep % |
11.531 |
13.918 |
-2.387 |
-17.1% |
| REM % |
23.180 |
23.502 |
-0.322 |
-1.4% |
| Awake min |
10.373 |
9.400 |
+0.973 |
+10.4% |
| Sleep min |
366.851 |
372.783 |
-5.932 |
-1.6% |
| FL_P95 |
0.089 |
0.093 |
-0.004 |
-4.7% |
| Min_Above_010 |
15.229 |
16.657 |
-1.429 |
-8.6% |
Recent 10-Night Context
| Date |
AHI |
OA |
CA |
PAvg |
P90 |
LAvg |
LMax |
L90 |
FL_P95 |
MinAbv010 |
MinAbv020 |
Deep% |
REM% |
Awake |
| 2026-09-04 |
3.37 |
16 |
0 |
7.37 |
10.22 |
0.31 |
28.8 |
0.0 |
0.07 |
11.4 |
2.4 |
14.66 |
22.41 |
6.5 |
| 2026-09-05 |
2.66 |
11 |
1 |
7.59 |
10.26 |
0.79 |
19.2 |
2.4 |
0.11 |
19.8 |
7.4 |
5.35 |
25.35 |
13.5 |
| 2026-09-06 |
1.40 |
7 |
1 |
6.72 |
9.44 |
0.27 |
14.4 |
0.0 |
0.08 |
14.4 |
3.8 |
17.64 |
19.40 |
10.0 |
| 2026-09-07 |
1.85 |
9 |
1 |
7.23 |
10.40 |
1.21 |
21.6 |
3.6 |
0.11 |
21.3 |
2.2 |
8.05 |
22.47 |
13.0 |
| 2026-09-08 |
2.25 |
9 |
0 |
7.38 |
9.34 |
1.12 |
18.0 |
3.6 |
0.09 |
15.0 |
3.7 |
7.85 |
19.22 |
11.0 |
| 2026-09-09 |
1.40 |
7 |
1 |
7.68 |
10.52 |
0.63 |
13.2 |
1.2 |
0.11 |
20.5 |
4.8 |
15.10 |
29.01 |
2.5 |
| 2026-09-10 |
2.10 |
10 |
3 |
7.14 |
9.30 |
0.14 |
7.2 |
0.0 |
0.07 |
9.7 |
1.4 |
11.44 |
20.34 |
7.0 |
| 2026-09-11 |
2.53 |
13 |
0 |
6.57 |
9.52 |
0.62 |
24.0 |
0.0 |
0.05 |
8.9 |
2.4 |
11.97 |
21.87 |
10.5 |
| 2026-09-12 |
1.81 |
8 |
1 |
7.17 |
9.92 |
0.45 |
14.4 |
1.2 |
0.08 |
12.7 |
3.9 |
18.06 |
27.63 |
9.5 |
| 2026-09-13 |
2.60 |
14 |
1 |
7.78 |
10.16 |
0.47 |
18.0 |
0.0 |
0.11 |
19.1 |
7.6 |
8.29 |
20.78 |
18.1 |
CPAP → Sleep-Stage Correlations (all 74 matched nights)
Correlations computed across the full available history (2026-06-23 → 2026-09-13).
| Pair |
the user r |
Interpretation |
| AHI vs deep_pct |
+0.030 |
negligible |
| AHI vs awake_min |
+0.068 |
negligible |
| OA vs awake_min |
+0.099 |
very weak positive |
| CA vs deep_pct |
+0.032 |
negligible |
| Leak_Avg vs deep_pct |
-0.182 |
weak negative — higher leak nights show slightly less deep sleep |
| Leak_90 vs deep_pct |
-0.186 |
weak negative — same direction |
| Leak_Max vs awake_min |
+0.085 |
negligible |
| FL_P95 vs awake_min |
-0.098 |
negligible |
| Min_Above_010 vs deep_pct |
+0.011 |
negligible |
Bucket comparison (Leak_90 median split, median = 2.40):
- Low-leak nights (n=36): mean deep% 15.69, awake 14.69 min
- High-leak nights (n=38): mean deep% 14.27, awake 14.12 min
- Δ (high – low): deep% -1.41, awake -0.58 min
The leak → deep-sleep weak negative signal persists but is essentially unchanged from prior reports. Notably, tonight breaks the leak/deep pattern in a striking way: Leak_Avg 0.47 (recent low) yet Deep% 8.29% (recent-window near-low). The whole set of CPAP-side metrics has poor explanatory power for tonight's sleep-stage collapse — the driver is likely non-CPAP (position, congestion, alcohol, evening food, illness, allergen exposure, or something else physiological).
Tonight's data pattern: OA count and Min_Above_020 are both at recent-window highs, and the machine's Pressure_Avg pushed to its highest since 09-04, all while leak stayed excellent. This suggests upper-airway obstruction/collapsibility was elevated tonight for reasons the CPAP data cannot see — not a mask/seal issue.
Caveats
- Sleep session started at 23:45:58 on 2026-09-13 (before midnight) — matched to the 2026-09-13 CPAP file per the standard convention.
- Ignored the daytime 29-min session at 06:09 on 2026-09-13 (a brief nap earlier that day, not the primary night's sleep).
- Apple Watch SpO2/pulse channels not present in current CPAP CSV export.
- Correlations are directional/small-effect only — the tonight-vs-baseline architectural collapse is much larger than what CPAP-side correlations alone would predict.
- All flagged metrics remain within clinically safe ranges (AHI < 5); the flags mark deviations from the user's own recent baseline, not clinical thresholds.
- The
health-report-2026-07-18-to-2026-07-19 2.json duplicate file was again ignored in favor of the primary copy.
- Correlation matched-night count dropped to 74 (from 78 last report) — some earlier nights in the dataset were re-evaluated with the same matching rule and a few borderline daytime-session matches were excluded.