The Saint-Hyacinthe ER occupancy rate, worked out by hand
Santé Québec publishes two numbers every hour for the Hôpital Honoré-Mercier emergency room: an occupancy rate as a percentage, and a count of people present. This site keeps them hour by hour; the ministry shows only the latest hour. The page below shows where the percentage comes from, with every operation written out.
The divisor
The occupancy rate is a count of occupied stretchers divided by a count of stretchers, times one hundred. The divisor is written nowhere on the dashboard. So I went looking for it in the readings themselves, and it is in there.
Over 163 consecutive hours, from 3 September 21:00 to 10 September 15:00 2026, the ER published a rate in every one of those hours. Those 163 readings contain only 24 different values, from 73 % to 162 %. Between neighbouring values the gap is 4 points, except in three places where it is 3. A rate that advances only in regular jumps has a small whole number underneath it, and the mean gap, 3.87 points, says which one: around 26.
The 24 values, one by one
Left column: the published value. Middle column: that value times 26 divided by 100, rounded — 73 gives 18.98, so 19 stretchers. Right column: that stretcher count turned back into a percentage, stretchers × 100 ÷ 26, rounded as well. The right column reproduces the left column for all 24 values. I ran the same test with 24, 25, 27 and 28 underneath: they land on 12, 7, 7 and 6 values out of 24. Only 26 gives all of them back.
| Published rate | Occupied stretchers (rate × 26 ÷ 100) | Back to the rate (stretchers × 100 ÷ 26) |
|---|---|---|
| 73 % | 19 | 73 % |
| 77 % | 20 | 77 % |
| 81 % | 21 | 81 % |
| 85 % | 22 | 85 % |
| 88 % | 23 | 88 % |
| 92 % | 24 | 92 % |
| 96 % | 25 | 96 % |
| 100 % | 26 | 100 % |
| 104 % | 27 | 104 % |
| 108 % | 28 | 108 % |
| 112 % | 29 | 112 % |
| 115 % | 30 | 115 % |
| 119 % | 31 | 119 % |
| 123 % | 32 | 123 % |
| 127 % | 33 | 127 % |
| 131 % | 34 | 131 % |
| 135 % | 35 | 135 % |
| 138 % | 36 | 138 % |
| 142 % | 37 | 142 % |
| 146 % | 38 | 146 % |
| 150 % | 39 | 150 % |
| 154 % | 40 | 154 % |
| 158 % | 41 | 158 % |
| 162 % | 42 | 162 % |
No intermediate value appears in 163 hours: nothing between 131 and 135, nothing between 96 and 100. The 162 % published on 9 September at 10:00 therefore reads as 42 occupied stretchers out of 26: 42 × 100 ÷ 26 = 161.5, rounded to 162. The 131 % I quoted in my first email works out at 34 stretchers.
The second number
The count of people present is published separately and divides nothing. A person sitting in the waiting room figures among those present without figuring in the rate, having no stretcher. The ratio present ÷ 26 can therefore be computed directly.
- Mean over the 163 hours: 2.33 people present per licensed stretcher.
- Minimum: 1.54. Maximum: 3.23, reached 10 September at 15:00 with 84 people present (84 ÷ 26 = 3.23).
- 128 of the 163 hours have at least two people present per licensed stretcher.
On 10 September at 15:00 the ER published 135 % and 84 people present, 34 of whom were still waiting to see a physician. The 135 % describes 35 stretchers.
Redoing the calculation
The Hôpital Honoré-Mercier permit number is 51229193. The address https://sante.handled.tools/api/facility/51229193/history?hours=250 returns the hourly readings as JSON: timestamp ts, rate occ_pct, people present present, awaiting triage-to-physician waiting. To redo the path: sort the distinct occ_pct values, look at the gaps, then try a denominator and see whether it gives every value back. I will send the 163 hours as a CSV if you would rather work in a spreadsheet.
One limit: nothing in the data says whether all 26 stretchers were open during the week. An ER can close some without its licence changing, and the rate would go on dividing by 26. Santé Québec Montérégie-Est, which operates the Hôpital Honoré-Mercier, is who to ask for the number actually open. The CISSS des Laurentides, asked on 6 September about its own ERs, answered on 10 September that the percentage there is computed on licensed stretchers; that is a different establishment, and I cite it as an indication for Saint-Hyacinthe.
Page published 10 September 2026. Window: 3 September 21:00 – 10 September 15:00 2026, 163 hourly readings, no gaps.