Sources: health-report-2026-07-24-to-2026-07-25.json (Apple Watch sleep stages, exported 08:19 on 07-25) + OSCAR_Greg_by_Day_2026-07-24.csv (ResMed AirSense/AirFlow 11 via OSCAR). Matched by calendar date of sleep-session start (23:11 on 07-24).
AHI jumped to 2.34, breaking a 3-night streak of AHI ≤ 1.6 (0.88, 1.59, 1.29 on 07-21–07-23). The rise is driven almost entirely by obstructive apneas: OA count = 15, more than double every night in the trailing week (range 2–12) and the highest OA count since 07-11 (21). Hypopneas were 0 and central apneas just 1, so this looks like a single-night obstructive event cluster rather than a broader therapy shift. Still well below the late-June peak (OA 23–29/night), and trailing 7-day AHI (1.53) is barely changed from last night's 1.59 — but worth watching whether 07-25 reverts or continues elevated.
Leak 90th percentile dropped to 0.0 L/min, resolving the two-night streak at 2.4 L/min flagged in the prior report. This confirms the 07-21–07-22 elevated-leak episode was transient and the mask seal has self-corrected — no cushion inspection needed per the prior report's decision rule. Leak avg held flat at 0.48 (unchanged from 07-23), and leak max spiked to 21.6 L/min (a brief episode) without affecting the 90th-pct or average, consistent with a short positional shift rather than a sustained seal problem.
Awake time rose to 24.6 min — roughly double the prior night (13.0 min) and above the trailing 7-day average (20.9 min). Deep sleep (16.7%) landed almost exactly on the trailing 7-day average (16.5%), matching the prior report's prediction of reversion toward the 12–16% baseline after two elevated-deep nights (07-21: 28.4%, 07-23: 25.5%). Nothing alarming here, but the awake-time increase tracks with tonight's AHI spike, consistent with the dataset's weak positive AHI-vs-awake relationship (see below).
Pressure avg continued its multi-week decline to 7.26 cmH₂O — a new low in the tracked dataset (prior trend: 8.49 → 8.10 → 8.00 → 7.88 → 7.26), even as AHI rose tonight. Worth noting the machine is running at lower pressure than any prior night while obstructive events picked up — not necessarily causal, but a pattern to keep an eye on if OA counts stay elevated.
| Metric | Value |
|---|---|
| Bedtime (session start) | 23:11 |
| Time in bed | 6.72 h (402.9 min) |
| Total sleep duration | 6.31 h (378.3 min) |
| Sleep efficiency (TST/TIB) | 93.9% |
| Awake time | 24.6 min ↑ (vs 13.0 min prior night) |
| Core sleep | 3.71 h (222.3 min) — 58.8% |
| Deep sleep | 1.05 h (63.2 min) — 16.7% |
| REM sleep | 1.55 h (92.8 min) — 24.5% |
| CPAP usage | 6.85 h |
| AHI | 2.34 ↑ (vs 1.29 prior night) |
| RDI | 2.34 |
| OA / H / CA / UA / RERA | 15 / 0 / 1 / 0 / 0 |
| Pressure avg | 7.26 cmH₂O (new low) |
| Pressure 90th %ile | 10.6 cmH₂O |
| Pressure max | 12.9 cmH₂O |
| Leak avg | 0.48 L/min (unchanged) |
| Leak max | 21.6 L/min |
| Leak 90th %ile | 0.0 L/min ↓ (vs 2.4 for two prior nights) |
| Date | AHI | OA | RERA | Leak avg | Leak 90th | Deep % | Awake (min) | Efficiency |
|---|---|---|---|---|---|---|---|---|
| 07-17 | 2.78 | 12 | 0 | 0.27 | 1.2 | n/a* | n/a* | n/a* |
| 07-18 | 2.35 | 10 | 0 | 0.54 | 0 | 14.7% | 36.6 | 90.3% |
| 07-19 | 1.29 | 6 | 1 | 0.92 | 0 | 10.1% | 14.0 | 96.2% |
| 07-20 | 0.94 | 2 | 0 | 0.03 | 0 | 8.3% | 21.1 | 94.4% |
| 07-21 | 0.88 | 4 | 1 | 0.85 | 1.2 | 28.4% | 27.1 | 93.3% |
| 07-22 | 1.59 | 7 | 1 | 1.08 | 2.4 | 12.0% | 10.0 | 97.5% |
| 07-23 | 1.29 | 7 | 0 | 0.48 | 2.4 | 25.5% | 13.0 | 96.8% |
| 07-24 | 2.34 | 15 | 0 | 0.48 | 0.0 | 16.7% | 24.6 | 93.9% |
*07-17 has no matched Apple Watch sleep-stage session in the current export set (see caveats).
Split at leak ≤ 0.30 vs. > 0.30 L/min:
the user r (leak avg vs. deep %): −0.29 (n=33) — a weak but now directionally consistent negative relationship (higher leak, somewhat less deep sleep), clearer than the +0.13 seen in the prior report once more nights are included. the user r (leak avg vs. awake min): +0.09 — still close to zero. the user r (AHI vs. awake min): +0.22 — weak positive, consistent with tonight's pattern (AHI up, awake time up). the user r (AHI vs. deep %): +0.03 — no relationship.
Interpretation: Across the full matched history, leak and deep sleep show a modest negative association, and AHI and awake time show a modest positive one — both directionally sensible but weak (|r| < 0.3), so treat as tendencies rather than reliable predictors on any single night. Tonight's combination (leak resolved, AHI up, awake time up) fits the AHI-awake relationship better than any leak-driven story.
| Metric | Tonight (07-24) | 7-day avg (07-18–07-24) | Prior 7-day avg (07-11–07-17) | 30-day avg (06-25–07-24) |
|---|---|---|---|---|
| AHI | 2.34 | 1.53 | 2.75 | 2.68 |
| Leak avg (L/min) | 0.48 | 0.63 | 0.33 | 0.94 |
| Pressure avg (cmH₂O) | 7.26 | 7.78 | 8.90 | 8.51 |
| Usage hours | 6.85 | 6.64 | 6.48 | 6.57 |
| Deep sleep % | 16.7% | 16.5% | 15.0% | 16.5% |
| REM sleep % | 24.5% | 23.9% | 23.4% | 23.7% |
| Awake time (min) | 24.6 | 20.9 | 14.1 | 16.7 |
Despite tonight's spike, trailing 7-day AHI (1.53) is essentially flat vs. yesterday's trailing figure (1.59) and remains well below both the prior-week (2.75) and 30-day (2.68) averages — one elevated night hasn't reversed the multi-week improvement. Pressure avg continues its steady decline (8.90 → 8.51 → 7.78 → 7.26 across the comparison windows).
| 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 |
| 07-17 to 07-23 (prior trailing 7d) | 1.59 | 0.60 | 7 |
| 07-18 to 07-24 (current trailing 7d) | 1.53 | 0.63 | 7 |
The multi-week downtrend in AHI remains intact — this is the second-lowest trailing-7-day AHI in the tracked history, just behind yesterday's 1.59.
Primary watch item: whether 07-24's OA spike (15) repeats on 07-25. If it does, and pressure avg stays at/below 7.3, that combination (lower pressure + elevated obstructive events) is worth flagging as a possible undertreatment signal. If 07-25 returns to the 4–12 OA range typical of the past week, tonight looks like an isolated event.
Leak: Resolved. No action needed — the 07-21–07-22 elevated-leak episode is confirmed transient (90th pct now 0.0 for the first time in the dataset).
Awake time / deep sleep: Both landed close to trailing averages; no concerning pattern here independent of the AHI question above.