Sources: health-report-2026-08-16-to-2026-08-17.json (Apple Watch sleep stages, exported 06:05 on 08-17) + OSCAR_Greg_by_Day_2026-08-16.csv (ResMed AirSense/AirFlow 11 via OSCAR). Session start was 23:02 on 08-16 — a normal, on-schedule bedtime after the very-late 00:07 start on 08-15. CPAP hours 6.22 vs. Watch TIB 6.06 (Δ ≈ 9 min, tightest alignment of the last 10 nights).
| Metric | Value |
|---|---|
| Bedtime (session start) | 23:02 on 08-16 (normal) |
| Time in bed | 6.06 h (363.8 min) |
| Total sleep duration | 5.96 h (357.8 min) |
| Sleep efficiency (TST/TIB) | 98.34% — highest of last 10 nights; above 7-day (94.7%) and 30-day (95.1%) |
| Awake time | 6.0 min — rank 56/61 (top-9% lowest); −72% vs. 7-day (21.1), −70% vs. 30-day (19.7) |
| Core sleep | 3.91 h (234.3 min) — 65.50% |
| Deep sleep | 0.88 h (52.7 min) — 14.73% — near-exactly the 30-day mean (15.11%) |
| REM sleep | 1.18 h (70.7 min) — 19.78% — below 30-day (24.55%) but within normal variance |
| CPAP usage | 6.22 h — near 7-day (6.61), −7% vs. 30-day (6.73) |
| AHI | 1.13 — lowest of the last 10 nights; −32% vs. 7-day (1.66), −35% vs. 30-day (1.74) |
| RDI | 1.45 (RERA = 2) |
| OA / H / CA / UA / RERA | 3 / 3 / 1 / 0 / 2 — OA broke the 5-in-a-row ≥ 8 streak; H slightly elevated but low absolute |
| OA per hour | 0.48 /h — well below trailing 30-day OA-rate (~1.18/h); lowest since 08-10 |
| Pressure avg | 7.92 cmH₂O — pulled back from 08-15's 8.35; +0.40 vs. 30-day (7.52) |
| Pressure 90th %ile | 10.94 cmH₂O — the highest of the last 10 nights, tied with 08-09 |
| Leak avg | 0.91 L/min — ends the 6-night clean streak; +0.60 vs. 08-15 |
| Leak 90th %ile | 4.8 L/min — highest 90th since 08-09; but still in "moderate" not "very-high" bucket |
| Compliance flag | 1 (compliant) |
| Date | AHI | OA | H | CA | RERA | Leak avg | P_avg | P_90 | Hours | Deep % | REM % | Awake (min) | Eff |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 08-07 | 1.69 | 8 | 2 | 1 | 1 | 1.20 | 7.39 | 9.66 | 6.50 | 20.53 | 31.47 | 6.5 | 98.30% |
| 08-08 | 2.34 | 6 | 10 | 0 | 0 | 0.87 | 7.63 | 9.90 | 6.83 | 16.17 | 26.60 | 15.6 | 96.12% |
| 08-09 | 1.34 | 6 | 3 | 0 | 2 | 3.00 | 7.79 | 10.94 | 6.72 | 10.97 | 23.65 | 15.1 | 96.19% |
| 08-10 | 0.58 | 3 | 0 | 0 | 0 | 0.17 | 6.89 | 9.46 | 5.20 | 15.30 | 20.31 | 20.1 | 94.73% |
| 08-11 | 2.57 | 12 | 4 | 0 | 0 | 0.42 | 7.28 | 10.02 | 6.62 | 13.57 | 25.24 | 21.1 | 94.61% |
| 08-12 | 2.68 | 13 | 3 | 2 | 1 | 0.28 | 7.99 | 10.00 | 6.72 | 20.78 | 19.11 | 28.1 | 92.80% |
| 08-13 | 1.29 | 8 | 0 | 2 | 0 | 0.20 | 7.01 | 9.54 | 7.75 | 10.93 | 26.70 | 20.1 | 95.60% |
| 08-14 | 1.63 | 11 | 1 | 0 | 1 | 0.22 | 7.94 | 10.86 | 7.38 | 9.73 | 29.20 | 33.6 | 92.19% |
| 08-15 | 1.53 | 8 | 1 | 0 | 0 | 0.31 | 8.35 | 10.38 | 5.88 | 26.72 | 24.89 | 10.0 | 97.04% |
| 08-16 | 1.13 | 3 | 3 | 1 | 2 | 0.91 | 7.92 | 10.94 | 6.22 | 14.73 | 19.78 | 6.0 | 98.34% |
Six-night OA composite (08-11 → 08-16): 55 OA + 9 H + 5 CA + 3 RERA across 40.57 h of use = 1.36 OA/h overall. If tonight is subtracted, the 5-night 08-11 → 08-15 composite is 52 OA over 34.35 h = 1.51 OA/h. Tonight materially pulls the trailing rate down (only 3 OA in 6.22 h). The cluster's momentum has broken; the next 2–3 nights determine whether this was a genuine reset or a one-night dip.
Buckets on nightly leak average:
the user r (leak avg vs. deep %): −0.25 (n=60). the user r (leak 90th vs. deep %): −0.26. the user r (leak avg vs. awake min): −0.06. the user r (AHI vs. deep %): +0.03. the user r (AHI vs. awake min): +0.05. the user r (OA vs. awake min): +0.11. the user r (pressure avg vs. deep %): −0.18. the user r (pressure 90th vs. deep %): −0.19.
Interpretation: Tonight is a rare clean-respiratory / moderate-leak / normal-architecture night — the CPAP side is uniformly good (AHI 1.13, OA 3, P_avg pulled back) and the sleep side is uniformly good on the objective consolidation metrics (awake 6.0 min, efficiency 98.34%). The moderate leak uptick (0.91 avg) is the only slight negative and did not translate into any measurable disruption. This is the cleanest single-night reading in about a week.
The prior report predicted that 08-15's top-3-ever deep % (26.72% on TIB 5.65 h) was a duration artifact — that a normal-length night would show a deep % much closer to the 20–21% range or lower. Tonight (TIB 6.06 h, deep 14.73%) is a direct confirmation: the short-window compression fully resolved when the sleep window normalized, and deep % returned to the trailing mean. This retrospectively strengthens the "duration, not resolution" reading of 08-15's architecture.
| Metric | Tonight (08-16) | 7-day avg (08-09–08-15) | 30-day avg (through 08-15) |
|---|---|---|---|
| AHI | 1.13 | 1.66 | 1.74 |
| OA count | 3 | 8.7 | 7.9 |
| OA per hour | 0.48 /h | ~1.32 /h | ~1.18 /h |
| Leak avg (L/min) | 0.91 | 0.66 | 0.76 |
| Leak 90th (L/min) | 4.8 | 2.23 | 2.28 |
| Pressure avg (cmH₂O) | 7.92 | 7.61 | 7.52 |
| Pressure 90th (cmH₂O) | 10.94 | 10.17 | 10.05 |
| Usage hours | 6.22 | 6.61 | 6.73 |
| Deep sleep % | 14.73% | 15.43% | 15.11% |
| REM sleep % | 19.78% | 24.16% | 24.55% |
| Awake time (min) | 6.0 | 21.1 | 19.7 |
| Efficiency | 98.34% | 94.65% | 95.10% |
| TIB (h) | 6.06 | 6.62 | 6.63 |
Verdict: tonight is the best night on the CPAP-respiratory side in ~2 weeks and the best sleep-consolidation night since 08-07, achieved on a normal-length window. The one caveat is REM % below the trailing average and a moderate leak uptick — neither is alarming, but if either persists it's worth watching.
| Week (Wed–Tue) | AHI avg | Leak avg | P_avg | P_90 avg | OA avg | Deep % | Awake (min) | Hours | n |
|---|---|---|---|---|---|---|---|---|---|
| 07-15 to 07-21 | 1.82 | 0.43 | 8.27 | 10.88 | 7.6 | 15.95% | 17.6 | 6.42 | 7 |
| 07-22 to 07-28 | 1.68 | 0.87 | 7.41 | 10.22 | 8.6 | 16.30% | 17.4 | 6.76 | 7 |
| 07-29 to 08-04 | 1.74 | 0.70 | 7.03 | 9.27 | 7.3 | 12.60% | 24.7 | 7.12 | 7 |
| 08-05 to 08-11 | 1.83 | 1.16 | 7.51 | 10.13 | 7.6 | 14.83% | 14.0 | 6.37 | 7 |
| 08-12 to 08-18 (partial, n=5) | 1.65 | 0.38 | 7.84 | 10.34 | 8.6 | 16.58% | 19.6 | 6.79 | 5 |
Partial-week update: (1) weekly OA average slipped from 10.0 (n=4) to 8.6 (n=5) with tonight's 3 pulling the average down — the weekly OA is now back inside the trailing 4-week range (7.3–8.6). (2) Weekly AHI 1.65 — lowest partial-week reading in ~5 weeks. (3) Weekly leak avg 0.38 still the lowest weekly leak in the dataset by a wide margin, even after tonight's uptick. (4) Deep % weekly 16.58% is normal; the two extreme readings (26.72 on 08-15, 9.73 on 08-14) largely cancel.
The 5-week AHI trend (1.82 → 1.68 → 1.74 → 1.83 → 1.65) is now bending down. OA weekly average, having been the informative "load" signal, has also come off its peak. If the next 2–3 nights hold OA ≤ ~6, the "sustained obstructive cluster" story from 08-11 → 08-15 will look like a discrete episode rather than a regime change.
The obstructive cluster's momentum has broken. For 5 consecutive nights (08-11 → 08-15) OA was ≥ 8 and the auto-titrator escalated pressure by +1.34 cmH₂O (7.01 → 7.94 → 8.35). Tonight OA = 3, P_avg pulled back to 7.92 (−0.43 cmH₂O), and AHI hit 1.13 — the lowest single-night AHI since 08-10. This is what a genuine reset looks like on the CPAP side: fewer events and the machine releasing pressure that it had been holding to chase load. The 08-15 watch-condition — "if OA ≥ 8 again, this becomes 6-of-6 and warrants a manual minimum-pressure raise" — did not trigger. The case for manual intervention is weaker now than it was 24 hours ago; give it 2–3 more nights before drawing a conclusion.
The 08-15 deep-sleep bounce was, as predicted, a short-window artifact. Tonight's normal-length night (TIB 6.06 h) shows deep sleep back at 14.73%, essentially exactly the 30-day mean. This is diagnostically valuable: it confirms the underlying architecture profile of the last 2–3 weeks (deep % in the low-teens on 08-13/08-14, mid-teens on 08-16) has not changed. The 08-13/08-14 low-deep readings are the baseline, and 08-15's number was a compression effect. Whether the low-deep baseline is "the new normal" or will recover to the mid-teens–20% range remains the open question — one more full-length night in the next 2–3 days should clarify.
Leak channel — first noise in a week. Six clean nights (avg 0.27 L/min) ended tonight at 0.91 avg / 4.8 90th. Not alarming, but a shift. On the same night, respiratory events dropped sharply — so the leak uptick was not causing disruption. Two possibilities to differentiate over the next few nights: (a) transient mask position drift that will self-correct, or (b) the start of a slow leak regression. If leak avg is > 0.5 for 2+ of the next 3 nights, the clean streak is genuinely over; if it drops back under 0.3 tomorrow, the uptick was one-night noise.
What to watch next. (1) OA on 08-17 — if OA stays ≤ 6 and P_avg stays ≤ ~7.9, the obstructive cluster is resolved. If OA rebounds to ≥ 8, treat 08-16 as a single-night dip in an ongoing cluster and reconsider the minimum-pressure question. (2) Leak on 08-17 — whether tonight's 0.91 was noise or drift. (3) Deep % on the next normal-length night — 08-16's 14.73% suggests the baseline is around 15%; another reading below 12% would signal ongoing fragmentation. (4) Bedtime consistency — 23:02 tonight is a healthy correction from 00:07 on 08-15. Holding a 23:00–23:30 window through the week protects TIB.