🟢 Leak profile FULLY REVERTED from yesterday's records — Leak_90 0.0 is the window FLOOR (tied rank 24/26, 1st-lowest; z = -1.24 vs 30d). Yesterday's 26-night window records (Leak_Max 40.80, Leak_Avg 2.18, Leak_90 8.40) collapse tonight: Leak_Max 16.80 (rank 21/26, z = -0.72, -27.3% vs 30d), Leak_Avg 0.32 (rank 25/26 — 2nd-lowest in window, z = -1.25, -74.1% vs 30d), Leak_90 0.00 (window floor, z = -1.24, -100.0% vs 30d). The night-over-night deltas: Leak_90 -100.0% (8.40 → 0.00), Leak_Avg -85.3% (2.18 → 0.32), Leak_Max -58.8% (40.80 → 16.80). This is the single sharpest leak-profile reversion in the 26-night window; the "2-of-2 nights of elevated leak-dimension" pattern flagged yesterday was NOT the start of a sustained trend — it was a one-night spike bracketed by two clean nights (10-11 Leak_90 9.6 was similarly a one-off). No mask-fit intervention is required by this reading alone; whatever happened on 10-12 (position, cushion seat, momentary dislodgement) was NOT structural.
🟢 FL distribution ALSO reverted — Min_Above_010 10.2 min is the window FLOOR (rank 26/26, 1st-lowest; z = -1.48, -42.9% vs 30d). FL_P95 0.08 (rank 21/26, z = -0.97, -19.2% vs 30d), down from yesterday's record-tie 0.14 (-42.9% n/n). Min_Above_010 10.2 min (window FLOOR, z = -1.48, -42.9% vs 30d), down from yesterday's 26.7 min (-61.8% n/n). FL_Avg 0.012 (-42.9% vs 30d, z = -1.52). Min_Above_020 2.3 min (rank 23/26, -47.7% vs 30d). The 4-night FL_P95 pattern (10-10 → 10-13) is now 0.14 → 0.09 → 0.14 → 0.08 — the "alternating spike/reversion" pattern flagged in yesterday's report resolves as TWO isolated spikes (10-10, 10-12) separated by two clean floor nights (10-11, 10-13), not sustained near-record territory. Yesterday's "contradicts the single-night event verdict" conclusion was itself premature; the sample now suggests the user has 2 spike-prone nights per week, not a durable shift.
🟢 Pressure response reverted — P_Avg 7.13 is rank 23/26 (z = -1.21, -6.4% vs 30d); P_Max 11.26 is rank 23/26 (z = -0.89). Yesterday's 10-12 report described the 5-night pressure escalation (P_Avg 6.89 → 7.38 → 7.38 → 7.78 → 8.20) as "a durably higher operating pressure." Tonight that trend REVERSES: P_Avg 7.13 is -1.07 cmH2O (-13.0%) below yesterday's near-record, and lands in the window's bottom-4. The 6-night path is now 6.89 → 7.38 → 7.38 → 7.78 → 8.20 → 7.13 — the monotonic climb broke decisively. P_90 9.90 (rank 20/26) and P_Max 11.26 (rank 23/26, 4th-lowest) corroborate: the auto-titrator has pulled back from the plateau-based elevation. Combined with the leak reversion, this suggests yesterday's elevated-plateau pressure was a RESPONSE to the leak/FL load, not a structural shift — remove the stress inputs, and the pressure returns to baseline. This supports the auto-titrator's responsiveness.
🟢 Deep sleep REBOUND — 16.77% breaks the 5-night soft stretch (rank 6/26, z = +0.97, +27.6% vs 30d). The 10-08 → 10-12 pattern (9.40 → 14.02 → 10.27 → 8.59 → 11.04, mean 10.66%) extended to a 6th night tonight would have become a clear 2-week trend. Instead, 16.77% is the single highest Deep% in the last 7 nights and the 6th-highest in the 26-night window. Deep minutes 55.70 (vs 10-12's 36.13, +54.2% n/n) is the highest Deep-minutes reading since 2026-10-07 (69.77 min). The 6-night Deep% average pulls up to 11.68%, versus 30d 13.14 — still below baseline but no longer the near-floor trajectory. The hypothesis from yesterday's report (pressure-driven arousals suppressing deep%) gets a corroborating data point tonight: Pressure_Avg dropped from near-record 8.20 to bottom-4 7.13, and Deep% went from 11.04 to top-6 16.77. The 103-night Pressure_Avg-vs-Deep_pct correlation (now 105 nights) remains weak at the population level, so this is a one-night pairing, not a mechanistic claim — but the directional alignment is clean.
🟡 Awake time 17.06 min — rank 3/26 in the window (z = +1.25, +77.4% vs 30d); sleep efficiency 95.11% is 2nd-lowest in the window (rank 24/26, z = -1.45). This is the UNUSUAL feature of tonight. Normally a clean-leak / low-FL / low-pressure night correlates with clean sleep architecture (low awake, high efficiency). Tonight breaks the pattern: despite the cleanest leak and FL readings in weeks, awake time is 2x the 30d average, and sleep efficiency drops below 96% for the first time in the 7-night trailing window. Over the full 26-night window only 10-09 (10.54 min awake, 96.95% eff) and 10-04 (1.51 min awake, 99.63% eff) are the only two above 10 min awake in the window other than tonight. The in-sleep AHI/leak/FL profile does NOT explain this awake time — the arousal driver is something OTHER than respiratory stress. Candidate hypotheses: external disturbance (noise, bathroom trip, pet, partner movement), pre-sleep arousal state, or late-night anxiety/thought activity. The CPAP data cannot distinguish these; they are flagged as a prompt for subjective recall.
🟡 Short night again — TIB 349.24 min (5h 49.2m) — rank 23/26 (z = -1.07); usage hours 6.02 is rank 22/26. 2nd consecutive short night (10-12 TIB 335.68, 10-13 TIB 349.24). The 2-night mean 342.46 min is well below the 7d trailing 387.30. Sleep onset was 22:53:44 — the earliest sleep start of the week (10-07 → 10-13: 22:58:53, 22:14:34, 00:49:14, 22:38:55, 23:24:08, 23:18:17, 22:53:44) — so the short night came from a reasonable mix of early-ish start and early wake. Wake time was approximately 04:43 CDT. Combined with yesterday's ~04:54 wake, the pattern suggests an early-rise schedule, not a window artifact. Total sleep 332.18 min (5h 32.2m) — rank 25/26, 2nd-shortest in the window.
🟢 AHI 1.66 — 8th consecutive sub-2.0 AHI night (streak extends: 1.85 → 1.24 → 0.85 → 1.49 → 1.26 → 1.42 → 1.41 → 1.66). OA 8 (rank 8/26 in high direction — top-third of the window; z = -0.15 vs 30d is still below the 30d mean 8.63 but above the 7d avg 5.86 by +36.6%). H 2 (rank 17/26, z = -0.55); CA 0 (tied rank 7/26); RERA 0 (tied rank 15/26). Mild uptick in OA count tonight is NOT elevated versus baseline — the 7d trailing mean 5.86 is the comparison point, and tonight is +2.14 above it, but the 30d mean 8.63 is -0.63 above tonight. In the context of a clean leak+FL+pressure profile, these 8 OA events are "machine running well, catching the few residuals left." RDI = AHI (no RERAs), same clean scoring pattern as 10-12.
🟢 REM 24.02% — right at 30d baseline (z = +0.03, +0.4% vs 30d 23.93); REM minutes 79.78 is a modest recovery from yesterday's 69.24. REM% pattern of last 5 nights (10-09 → 10-13): 20.09 → 18.96 → 26.65 → 21.17 → 24.02 — the volatile-but-net-declining profile from yesterday's report has stabilized. 7d REM% average 23.10 → 23.23 with tonight included, approaching the 30d mean 23.93.
| Metric | Value |
|---|---|
| Sleep start | 2026-10-13T22:53:44-05:00 |
| Time in bed | 349.24 min (5h 49.2m) |
| Total sleep | 332.18 min (5h 32.2m) |
| Sleep efficiency | 95.11% |
| Awake time | 17.06 min |
| Core (N1+N2) sleep | 196.70 min / 59.21% |
| Deep (N3) sleep | 55.70 min / 16.77% |
| REM | 79.78 min / 24.02% |
| CPAP usage hours | 6.02 |
| Session count | 1 |
| AHI | 1.66 |
| RDI | 1.66 |
| Obstructive apneas (OA) | 8 |
| Hypopneas (H) | 2 |
| Central apneas (CA) | 0 |
| Unclassified apneas (UA) | 0 |
| RERA | 0 |
| Pressure_Avg | 7.13 |
| Pressure_Min | 4.00 |
| Pressure_Max | 11.26 |
| Pressure_90th | 9.90 |
| Leak_Avg | 0.32 |
| Leak_Max | 16.80 |
| Leak_90th | 0.00 |
| FL_Avg | 0.0120 |
| FL_P95 | 0.08 |
| FL_Max | 0.37 |
| Min_Above_005 (min) | 27.5 |
| Min_Above_010 (min) | 10.2 |
| Min_Above_020 (min) | 2.3 |
| Frac_NonZero (FL samples) | 21.88% |
| Metric | Tonight | 7d avg | 30d avg | Δ vs 7d | Δ vs 30d |
|---|---|---|---|---|---|
| AHI | 1.66 | 1.36 | 1.97 | +22.1% | -15.8% |
| RDI | 1.66 | 1.56 | 2.13 | +6.4% | -22.2% |
| OA count | 8.00 | 5.86 | 8.63 | +36.6% | -7.3% |
| H count | 2.00 | 2.86 | 3.40 | -30.0% | -41.2% |
| CA count | 0.00 | 0.14 | 0.50 | -100.0% | -100.0% |
| RERA count | 0.00 | 1.29 | 1.03 | -100.0% | -100.0% |
| Pressure_Avg | 7.13 | 7.50 | 7.62 | -4.9% | -6.4% |
| Pressure_90 | 9.90 | 10.22 | 10.20 | -3.1% | -3.0% |
| Pressure_Max | 11.26 | 12.43 | 12.10 | -9.4% | -6.9% |
| Leak_Avg | 0.32 | 1.52 | 1.23 | -78.9% | -74.1% |
| Leak_Max | 16.80 | 28.80 | 23.12 | -41.7% | -27.3% |
| Leak_90th | 0.00 | 5.14 | 4.24 | -100.0% | -100.0% |
| Usage hours | 6.02 | 6.56 | 6.42 | -8.3% | -6.2% |
| FL_P95 | 0.080 | 0.106 | 0.099 | -24.3% | -19.2% |
| FL_Avg | 0.0120 | 0.0171 | 0.0163 | -29.9% | -26.3% |
| FL_Max | 0.370 | 0.511 | 0.487 | -27.6% | -24.0% |
| Min_Above_005 | 27.5 | 42.7 | 40.0 | -35.6% | -31.3% |
| Min_Above_010 | 10.2 | 20.6 | 17.9 | -50.4% | -42.9% |
| Min_Above_020 | 2.3 | 5.66 | 4.40 | -59.4% | -47.7% |
| Frac_NonZero | 0.219 | 0.254 | 0.255 | -13.9% | -14.1% |
| Deep % | 16.77 | 11.95 | 13.14 | +40.3% | +27.6% |
| REM % | 24.02 | 23.10 | 23.93 | +4.0% | +0.4% |
| Core % | 59.21 | 64.95 | 62.94 | -8.8% | -5.9% |
| Awake min | 17.06 | 7.38 | 9.62 | +131.1% | +77.4% |
| TIB min | 349.24 | 387.30 | 380.23 | -9.8% | -8.2% |
| Total sleep min | 332.18 | 379.92 | 370.62 | -12.6% | -10.4% |
| Sleep efficiency | 95.11 | 98.07 | 97.47 | -3.0% | -2.4% |
| Metric | Recent 7d | Prior 7d | Δ | Δ % |
|---|---|---|---|---|
| AHI | 1.333 | 2.086 | -0.753 | -36.1% |
| RDI | 1.511 | 2.377 | -0.866 | -36.4% |
| OA | 5.857 | 8.857 | -3.000 | -33.9% |
| H | 2.571 | 4.000 | -1.429 | -35.7% |
| CA | 0.143 | 0.286 | -0.143 | -50.0% |
| RERA | 1.143 | 1.857 | -0.714 | -38.5% |
| Pressure_Avg | 7.493 | 7.651 | -0.159 | -2.1% |
| Pressure_90 | 10.174 | 10.257 | -0.083 | -0.8% |
| Pressure_Max | 12.300 | 11.746 | +0.554 | +4.7% |
| Leak_Avg | 1.416 | 1.553 | -0.137 | -8.8% |
| Leak_90 | 4.971 | 5.486 | -0.514 | -9.4% |
| Leak_Max | 26.914 | 22.114 | +4.800 | +21.7% |
| Usage hours | 6.494 | 6.454 | +0.040 | +0.6% |
| Deep % (Apple) | 12.614 | 13.924 | -1.310 | -9.4% |
| REM % (Apple) | 22.658 | 25.903 | -3.245 | -12.5% |
| Awake min (Apple) | 9.247 | 10.537 | -1.291 | -12.2% |
| Sleep min (Apple) | 372.965 | 373.038 | -0.073 | -0.0% |
| TIB min (Apple) | 382.212 | 383.575 | -1.363 | -0.4% |
| Sleep efficiency | 97.518 | 97.225 | +0.293 | +0.3% |
| FL_P95 | 0.106 | 0.091 | +0.014 | +15.6% |
| FL_Avg | 0.017 | 0.016 | +0.001 | +7.5% |
| Min_Above_010 | 19.886 | 15.929 | +3.957 | +24.8% |
| Min_Above_020 | 5.329 | 4.200 | +1.129 | +26.9% |
Note: the recent-7d means above include tonight's dramatic reversions on the leak side (L_Avg, L_90) but still show +21.7% Leak_Max and +15.6% FL_P95 week-over-week — because the recent 7d contains the two spike nights (10-10, 10-12) that pulled the mean up before tonight pulled it back. The 7d Deep% WoW -9.4% is narrower than yesterday's -15.2%: tonight's rebound shifted the recent 7d mean up by 0.66pp.
| Date | AHI | OA | CA | RERA | PAvg | P90 | LAvg | LMax | L90 | FL_P95 | MinAbv010 | MinAbv020 | Deep% | REM% | Awake |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2026-10-04 | 1.18 | 6 | 0 | 4 | 7.90 | 10.18 | 1.91 | 28.8 | 4.8 | 0.13 | 25.7 | 7.1 | 7.88 | 27.87 | 1.51 |
| 2026-10-05 | 5.20 | 21 | 0 | 4 | 7.93 | 12.36 | 1.76 | 22.8 | 7.2 | 0.09 | 15.7 | 8.2 | 15.92 | 25.21 | 8.03 |
| 2026-10-06 | 1.85 | 8 | 0 | 1 | 7.19 | 10.24 | 1.04 | 30.0 | 1.2 | 0.08 | 15.1 | 4.6 | 12.12 | 27.14 | 4.01 |
| 2026-10-07 | 1.24 | 7 | 0 | 0 | 7.69 | 10.72 | 1.50 | 34.8 | 3.6 | 0.12 | 23.3 | 4.6 | 18.21 | 23.43 | 6.02 |
| 2026-10-08 | 0.85 | 6 | 0 | 1 | 6.89 | 9.20 | 1.58 | 30.0 | 3.6 | 0.09 | 17.4 | 3.9 | 9.40 | 24.30 | 9.03 |
| 2026-10-09 | 1.49 | 6 | 1 | 5 | 7.38 | 9.88 | 1.34 | 24.0 | 4.8 | 0.08 | 14.0 | 6.7 | 14.02 | 20.09 | 10.54 |
| 2026-10-10 | 1.26 | 5 | 0 | 1 | 7.38 | 10.60 | 1.10 | 18.0 | 4.8 | 0.14 | 32.0 | 9.8 | 10.27 | 18.96 | 4.01 |
| 2026-10-11 | 1.42 | 4 | 0 | 1 | 7.78 | 10.18 | 1.89 | 24.0 | 9.6 | 0.09 | 15.6 | 4.1 | 8.59 | 26.65 | 9.53 |
| 2026-10-12 | 1.41 | 5 | 0 | 0 | 8.20 | 10.74 | 2.18 | 40.8 | 8.4 | 0.14 | 26.7 | 5.9 | 11.04 | 21.17 | 8.53 |
| 2026-10-13 | 1.66 | 8 | 0 | 0 | 7.13 | 9.90 | 0.32 | 16.8 | 0.0 | 0.08 | 10.2 | 2.3 | 16.77 | 24.02 | 17.06 |
Correlations computed across the full matched set after excluding pre-2026-06-23 low-confidence CPAP nights. The sample has grown by two nights (10-12, 10-13) since the 10-12 report's 103-night set.
| Pair | the user r | Interpretation |
|---|---|---|
| AHI vs deep_pct | +0.085 | negligible |
| AHI vs awake_min | +0.052 | negligible |
| OA vs awake_min | +0.109 | weak positive |
| CA vs deep_pct | +0.071 | negligible |
| Leak_Avg vs deep_pct | -0.162 | weak negative |
| Leak_90 vs deep_pct | -0.151 | weak negative |
| Leak_Max vs awake_min | +0.020 | negligible |
| FL_P95 vs awake_min | -0.093 | negligible |
| Min_Above_010 vs deep_pct | +0.006 | negligible |
Bucket comparison (Leak_90 median split, median = 2.40): - Low-leak nights (n=46): mean deep% 14.96, awake 14.21 min - High-leak nights (n=59): mean deep% 14.06, awake 12.20 min - Δ (high – low): deep% -0.90 pp, awake -2.01 min
Tonight's Leak_90 of 0.00 lands at the FLOOR of the low-leak bucket. Tonight's deep% of 16.77 is 1.81 points above the low-leak bucket mean (14.96) — consistent with the bucket's directional prediction. The 105-night population correlations are essentially unchanged from yesterday's 103-night set: Leak_Avg vs deep_pct remains the strongest signal (|r| = 0.16), MinAbv010 vs deep_pct remains at effectively zero. Tonight's low-leak/high-deep pairing is the EXPECTED outcome given the bucket means — unlike yesterday's high-leak/low-deep pairing, which was also directionally consistent. Two consecutive nights of leak-deep directional alignment is weak evidence for an individual-level leak-arousal mechanism, but the population-level |r| ceiling (~0.16) caps how much any single night can be attributed to the signal.