| Metric | Value |
|---|---|
| CPAP usage hours | 6.98 |
| Session count | 1 |
| AHI | 2.72 |
| RDI | 2.86 |
| Obstructive apneas (OA) | 13 |
| Hypopneas (H) | 6 |
| Central apneas (CA) | 0 |
| Unclassified apneas (UA) | 0 |
| RERA | 1 |
| Pressure_Avg | 7.66 |
| Pressure_Min | 4.00 |
| Pressure_Max | 13.52 |
| Pressure_90th | 10.50 |
| Leak_Avg | 2.37 |
| Leak_Max | 22.80 |
| Leak_90th | 9.60 |
| FL_Avg | 0.0137 |
| FL_P95 | 0.10 |
| FL_Max | 0.44 |
| Min_Above_005 | 34.8 |
| Min_Above_010 | 19.2 |
| Min_Above_020 | 5.1 |
| Frac_NonZero (FL samples) | 18.93% |
| Sleep-stage data | Not yet available (see caveats) |
Trailing windows exclude tonight by construction (they cover the 7 and 30 nights ending 2026-09-25).
| Metric | Tonight | 7d avg | 30d avg | Δ vs 7d | Δ vs 30d | 30d z |
|---|---|---|---|---|---|---|
| AHI | 2.72 | 2.41 | 2.23 | +12.9% | +21.9% | +0.74 |
| OA count | 13.00 | 8.86 | 10.23 | +46.8% | +27.1% | +0.79 |
| H count | 6.00 | 5.43 | 3.07 | +10.5% | +95.6% | +1.35 |
| CA count | 0.00 | 0.43 | 0.97 | -100.0% | -100.0% | -0.86 |
| RERA count | 1.00 | 0.57 | 0.73 | +75.0% | +36.4% | +0.29 |
| Pressure_Avg | 7.66 | 7.76 | 7.48 | -1.2% | +2.4% | +0.39 |
| Pressure_90th | 10.50 | 10.26 | 9.99 | +2.4% | +5.1% | +1.02 |
| Leak_Avg | 2.37 | 1.25 | 0.83 | +89.0% | +185% | +2.49 |
| Leak_Max | 22.80 | 24.34 | 18.64 | -6.3% | +22.3% | +0.56 |
| Leak_90th | 9.60 | 3.94 | 2.52 | +143.5% | +281% | +2.57 |
| Usage hours | 6.98 | 6.11 | 6.41 | +14.2% | +8.9% | +1.46 |
| FL_P95 | 0.10 | 0.10 | 0.09 | -3.0% | +6.4% | +0.29 |
| Min_Above_010 | 19.20 | 18.14 | 16.70 | +5.8% | +15.0% | +0.45 |
| Min_Above_020 | 5.10 | 3.56 | 3.71 | +43.4% | +37.4% | +0.78 |
| Metric | Recent 7d | Prior 7d | Δ | Δ % |
|---|---|---|---|---|
| AHI | 2.339 | 2.349 | -0.010 | -0.4% |
| OA | 8.571 | 11.857 | -3.286 | -27.7% |
| H | 5.571 | 2.000 | +3.571 | +178.6% |
| CA | 0.429 | 1.286 | -0.857 | -66.7% |
| Pressure_Avg | 7.766 | 7.430 | +0.336 | +4.5% |
| Pressure_90th | 10.380 | 9.791 | +0.589 | +6.0% |
| Leak_Avg | 1.534 | 0.467 | +1.067 | +228.4% |
| Leak_90 | 5.143 | 1.371 | +3.771 | +275.0% |
| Leak_Max | 25.543 | 14.571 | +10.971 | +75.3% |
| Usage hours | 6.227 | 6.464 | -0.237 | -3.7% |
| FL_P95 | 0.103 | 0.096 | +0.007 | +7.5% |
| Min_Above_010 | 18.486 | 16.400 | +2.086 | +12.7% |
| Min_Above_020 | 3.986 | 4.243 | -0.257 | -6.1% |
| Date | AHI | OA | H | CA | PAvg | P90 | LAvg | LMax | L90 | FL_P95 | MinAbv010 | MinAbv020 | Hours |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2026-09-17 | 2.41 | 11 | – | 2 | 7.17 | 9.22 | 0.52 | 10.80 | 1.20 | 0.07 | 12.0 | 3.3 | 6.22 |
| 2026-09-18 | 0.87 | 5 | – | 0 | 7.07 | 9.56 | 0.18 | 8.40 | 0.00 | 0.09 | 16.4 | 2.9 | 6.87 |
| 2026-09-19 | 3.23 | 15 | – | 0 | 7.60 | 9.64 | 0.41 | 14.40 | 1.20 | 0.10 | 16.8 | 2.1 | 6.18 |
| 2026-09-20 | 2.18 | 7 | – | 0 | 8.21 | 10.54 | 1.85 | 27.60 | 6.00 | 0.13 | 28.6 | 7.0 | 6.42 |
| 2026-09-21 | 1.94 | 5 | – | 1 | 7.82 | 10.46 | 0.90 | 14.40 | 1.20 | 0.09 | 16.2 | 2.9 | 6.20 |
| 2026-09-22 | 3.20 | 12 | – | 0 | 7.49 | 10.32 | 1.03 | 36.00 | 2.40 | 0.10 | 14.8 | 3.1 | 5.93 |
| 2026-09-23 | 1.94 | 9 | – | 0 | 7.78 | 10.06 | 2.87 | 26.40 | 12.00 | 0.10 | 17.4 | 3.3 | 6.18 |
| 2026-09-24 | 1.65 | 3 | – | 1 | 7.36 | 10.18 | 0.59 | 28.80 | 1.20 | 0.09 | 15.2 | 3.4 | 6.05 |
| 2026-09-25 | 2.74 | 11 | 4 | 1 | 8.04 | 10.60 | 1.13 | 22.80 | 3.60 | 0.11 | 18.0 | 3.1 | 5.83 |
| 2026-09-26 | 2.72 | 13 | 6 | 0 | 7.66 | 10.50 | 2.37 | 22.80 | 9.60 | 0.10 | 19.2 | 5.1 | 6.98 |
Correlations are unchanged from the 2026-09-25 report because tonight has no matched sleep-stage session yet (see caveats). Prior 71-night baseline (2026-07-13 → 2026-09-25):
| Pair | the user r | Interpretation |
|---|---|---|
| AHI vs deep_pct | -0.113 | weak negative |
| AHI vs awake_min | -0.183 | weak negative |
| OA vs awake_min | -0.054 | negligible |
| Leak_Avg vs deep_pct | -0.005 | negligible |
| Leak_90th vs deep_pct | -0.010 | negligible |
| Leak_Max vs awake_min | -0.201 | weak negative |
| FL_P95 vs awake_min | -0.063 | negligible |
| Min_Above_010 vs deep_pct | -0.180 | weak negative |
Bucket comparison from prior report (Leak_90 median split, median = 1.20): - Low-leak nights (n=37): mean deep% 14.25, awake 14.42 min - High-leak nights (n=34): mean deep% 13.97, awake 11.82 min
If tonight's Apple Watch session arrives later, it will be a high-leak, high-usage data point — the strongest test yet of whether elevated leak alone (in the absence of the short-TIB confound seen 2026-09-25) meaningfully suppresses deep sleep. Worth backfilling in the next report when the health export lands.
Tonight is essentially a mixed night: the biggest positive change since yesterday's report is duration — 6.98 h is a full hour longer than 2026-09-25 and unambiguously ends the 8-night declining-usage streak flagged in the last report. That was the "one metric to keep watching" from yesterday, and it moved in the right direction on the first night after the flag.
The concerning signal is leak. The last four reports have progressively flagged leak elevation and framed it as a "sustained weekly shift"; tonight is the strongest single-night confirmation. Leak_Avg 2.37 and Leak_90 9.60 both sit near +2.5 standard deviations from the 30-day baseline, and the recent-week-vs-prior-week comparison has now widened to +228%/+275%. The week-over-week pattern is no longer explainable as a single bad night; it is a level shift. Two mechanistic candidates worth naming: 1. Mask/cushion wear. Tonight's Leak_Max (22.80) is only mid-range — the elevation is not a runaway session, it's a persistently higher steady-state leak floor. That signature typically points to a cushion seal that has aged or a headgear tension that has crept looser, rather than a mid-night dislodgement. 2. Position or facial-hair change. Sustained higher steady-state leak can also follow from a bedtime-position shift (more side-sleeping into the pillow) or a beard/hair change. Nothing in the CPAP data can distinguish these from a cushion issue — worth a mental note if any of these have changed recently.
Event counts are also mildly elevated (AHI 2.72, OA 13, H 6) but still well within the user's own well-controlled range and far below the clinical threshold (AHI < 5). The H count in particular (6) is 2× the 30d baseline, and hypopneas often co-vary with leak because a leak-driven pressure delivery loss can shade a would-be-quiet breath toward partial obstruction. Not causal-proven, but consistent with the leak reading.
Pressure_90th at 10.50 is the auto-titrator responding, exactly as designed. Min_Above_020 at 5.1 min is elevated (+37% vs 30d) but Min_Above_005 (34.8) and Min_Above_010 (19.2) sit right on baseline — so the flow-limitation increase is concentrated at moderate levels, not at the top of the FL distribution.
Overall: a longer, leakier, slightly-more-eventful night than baseline. Zero clinical red flags. The single actionable observation continues to be inspect the mask cushion and headgear — the leak pattern has now been elevated for two straight weeks with the recent week markedly higher than the prior one.
health-report-*.json covers the night starting 2026-09-25 (session date 2026-09-26T00:05:36-05:00, matched to CPAP 2026-09-25). This report is CPAP-only; when the 2026-09-26-to-2026-09-27 health export arrives, the next report can backfill deep/REM/awake context and refresh the leak-vs-sleep-stage correlations.