Sources: health-report-2026-08-01-to-2026-08-02.json (Apple Watch sleep stages, exported 08:54 on 08-02) + OSCAR_Greg_by_Day_2026-08-01.csv (ResMed AirSense/AirFlow 11 via OSCAR). Matched by calendar date of sleep-session start (23:13 on 08-01).
This is the most disrupted night in the entire 47-night dataset by three separate measures. Awake time hit 66.7 minutes — more than double the previous worst (31.6 min on 06-25) and nearly 3x the trailing 7-day average (23.7 min). Sleep efficiency fell to 86.8%, the lowest on record (previous low: 91.2% on 06-25). Deep sleep dropped to 5.8%, also the lowest on record (previous low: 8.3% on 07-20) and barely a third of the 30-day average (16.5%). All three records were set on the same night.
This is the third straight night of rising awake time (07-30: 27.6 min → 07-31: 24.1 min → 08-01: 66.7 min), continuing the reversal flagged in yesterday's report. The trailing 7-day awake average has climbed from 18.6 min (as of 07-30) to 23.7 min (as of tonight).
Notably, this did not coincide with elevated CPAP events or leak. AHI was just 0.58 (lowest of the recent window) and leak was a moderate 0.4 L/min — both well within normal range. This points away from a CPAP-driven cause and toward something else: tonight also had the longest time in bed (8.45 h) and longest CPAP session (8.58 h) in the entire dataset, which mechanically allows more waking minutes to accumulate and would also dilute deep-sleep percentage (which concentrates earlier in a sleep period). Whether this reflects a longer, more fragmented night or a data/sensor artifact isn't distinguishable from this data alone.
| Metric | Value |
|---|---|
| Bedtime (session start) | 23:13 |
| Time in bed | 8.45 h (506.8 min) — longest in the dataset |
| Total sleep duration | 7.33 h (440.1 min) |
| Sleep efficiency (TST/TIB) | 86.8% — lowest in the dataset |
| Awake time | 66.7 min — highest in the dataset (2nd-highest: 31.6 min) |
| Core sleep | 6.13 h (367.8 min) — 83.6% — highest core % in the dataset |
| Deep sleep | 0.43 h (25.6 min) — 5.8% — lowest in the dataset |
| REM sleep | 0.78 h (46.7 min) — 10.6% — lowest since 07-19 |
| CPAP usage | 8.58 h — longest session in the dataset |
| AHI | 0.58 — lowest of the recent window |
| RDI | 0.58 |
| OA / H / CA / UA / RERA | 2 / 2 / 1 / 0 / 0 |
| Pressure avg | 6.33 cmH₂O — lowest in the dataset |
| Pressure 90th %ile | 8.14 cmH₂O |
| Leak avg | 0.4 L/min — moderate, unremarkable |
| Leak 90th %ile | 2.4 L/min |
| Date | AHI | OA | RERA | Leak avg | Leak 90th | Deep % | Awake (min) | Efficiency |
|---|---|---|---|---|---|---|---|---|
| 07-25 | 1.72 | 10 | 0 | 1.69 | 7.2 | 9.9% | 30.6 | 92.5% |
| 07-26 | 1.24 | 5 | 0 | 0.39 | 0.0 | 14.3% | 26.6 | 92.9% |
| 07-27 | 1.6 | 8 | 0 | 1.19 | 2.4 | 18.0% | 5.5 | 98.5% |
| 07-28 | 2.01 | 8 | 2 | 0.77 | 2.4 | 17.8% | 11.5 | 97.2% |
| 07-29 | 1.62 | 7 | 2 | 1.52 | 6.0 | 17.8% | 4.0 | 98.9% |
| 07-30 | 1.98 | 13 | 0 | 0.08 | 0.0 | 11.3% | 27.6 | 93.3% |
| 07-31 | 1.51 | 7 | 0 | 1.03 | 3.6 | 17.9% | 24.1 | 94.9% |
| 08-01 | 0.58 | 2 | 0 | 0.4 | 2.4 | 5.8% | 66.7 | 86.8% |
Split at leak ≤ 0.30 vs > 0.30 L/min (08-01's leak of 0.4 falls into the higher-leak bucket):
the user r (leak avg vs. deep %): −0.27 (n=41). the user r (leak avg vs. awake min): −0.01. the user r (AHI vs. awake min): −0.03. the user r (AHI vs. deep %): +0.04.
Interpretation: Tonight's extreme awake time (66.7 min) and rock-bottom deep% (5.8%) occurred with only moderate leak (0.4) and the lowest AHI in the recent window (0.58) — a night that sits far outside both correlation lines. Including tonight actually weakened the leak↔awake and AHI↔awake correlations toward zero (both were previously in the +0.07 to +0.20 range), because this outlier combines low leak/AHI with extreme disruption. This reinforces the developing picture across this dataset: neither leak nor AHI is a strong predictor of a bad night — something else (likely related to the unusually long time in bed) is the better explanation here.
| Metric | Tonight (08-01) | 7-day avg (07-26–08-01) | 30-day avg (07-03–08-01) |
|---|---|---|---|
| AHI | 0.58 | 1.51 | 2.06 |
| Leak avg (L/min) | 0.4 | 0.77 | 0.59 |
| Pressure avg (cmH₂O) | 6.33 | 6.93 | 7.99 |
| Usage hours | 8.58 | 7.07 | 6.69 |
| Deep sleep % | 5.8% | 14.7% | 16.5% |
| REM sleep % | 10.6% | 22.9% | 24.4% |
| Awake time (min) | 66.7 | 23.7 | 18.3 |
Every disruption-related metric is a clear outlier vs. both trailing windows: awake time nearly 3x the 7-day average and 3.6x the 30-day average; deep% at roughly a third of both baselines; REM% at less than half both baselines. Meanwhile AHI, leak, and pressure are all below their trailing averages — reinforcing that this was not a respiratory-event-driven disruption.
| Week | AHI avg | Leak avg | n |
|---|---|---|---|
| 06-22 to 06-28 | 2.84 | 1.98 | 7 |
| 06-29 to 07-05 | 3.24 | 1.41 | 7 |
| 07-06 to 07-12 | 3.00 | 0.43 | 7 |
| 07-13 to 07-19 | 2.22 | 0.42 | 7 |
| 07-20 to 07-26 | 1.43 | 0.71 | 7 |
| 07-27 to 08-02 (current, 7 of 7 logged) | 1.55 | 0.83 | 7 |
Trailing 7-day AHI (ending 08-01) is 1.51 — down from 1.79 a week earlier and comfortably the lowest trailing-week figure since this tracking began. AHI and leak are not the story this week; sleep architecture disruption is.
This is the single most notable night in the dataset and the clearest case yet for a cause other than CPAP performance. Every AHI/leak/pressure metric was better than average, yet awake time, deep%, and efficiency all hit their worst recorded levels. The most plausible mechanical explanation available in this data is time-in-bed: at 8.45 hours, tonight's sleep opportunity was almost an hour longer than the next-longest night, which both allows more absolute wake minutes to accumulate and dilutes deep-sleep percentage (deep sleep concentrates in the first sleep cycles). If a longer bedtime window recurs, worth watching whether the same pattern repeats — that would support the time-in-bed explanation over a one-off anomaly (illness, stress, alcohol, schedule disruption, etc., none of which are captured in this data).
Three consecutive nights of rising awake time (27.6 → 24.1 → 66.7 min) is now a real short-term trend, not just a single bad night. Worth watching whether 08-02 continues the climb or reverts toward the 07-27–07-29 lows.
AHI, pressure, and leak all continue to look good — 0.58 AHI is the best of the recent window, and pressure (6.33) and leak (0.4) are both unremarkable to favorable. The CPAP therapy itself does not appear to be the issue tonight.