Chasing Pressure Alarms in a 12-Room Airborne Infection Isolation Unit
A hospital's 12-room airborne infection isolation (AII) unit was commissioned six years ago to the facility's design basis: negative pressure relative to the corridor, 12 air changes per hour total, 2 outdoor air changes, and no recirculation within the room. Over the past four months the building automation system has logged 30-60 pressure alarms per week. Trend data show rooms holding roughly -0.010 in. w.c. overnight but drifting toward -0.001 to -0.003 in. w.c. between 06:00 and 09:00 and again at shift change, when corridor doors are propped open for long periods. Five months ago an energy contractor reprogrammed the exhaust fan VFDs to modulate on a duct static setpoint reset; two rooms share a common exhaust riser. Room pressure monitors were last calibrated 26 months ago, and last week's balometer readings put the rooms at 9-11 ACH.
1. The infection preventionist wants all 12 rooms taken out of service immediately. What do you recommend, and in what order do you investigate?
Read the worked answer
I would not take 12 rooms out of service on the basis of an alarm log alone, but I would not let an unverified room house an infectious patient either. The decision has to rest on measured pressure, not on a monitor that is two years past calibration.
Immediate (same day): verify each room with a calibrated handheld micromanometer and a visual smoke/puff test at the door under-closure, with the door closed and the room's exhaust running. Any room measuring positive or neutral relative to the corridor is removed from AII service and the patient relocated per infection prevention's direction. Rooms that verify negative stay in use with documented hourly checks until the trend issue is resolved.
Then, in order:
- Calibrate or replace the pressure monitors - 26 months since calibration makes the alarm flood itself suspect, and nuisance alarms erode staff response.
- Re-examine the VFD static reset. Modulating exhaust on duct static is the most likely real cause of the 06:00-09:00 drift; the unit's negative pressure is a function of exhaust-to-supply offset, not duct pressure.
- Check the shared exhaust riser for cross-talk between the paired rooms - one room's damper movement can pull the other off setpoint.
- Review door-management practice and add an alarm delay long enough (typically 30-60 seconds) to avoid nuisance trips from normal traffic without masking a real loss of pressure.
- Re-balance and re-verify ACH against the design basis, since measured 9-11 ACH is already below the commissioned 12.
2. The energy contractor argues restoring constant-volume exhaust will cost roughly $38,000 a year in fan energy. How do you frame the decision for the CFO?
Read the worked answer
I would frame it as a comparison of two costs, not as comfort versus sustainability.
Cost of the proposed 'savings.' $38,000 per year is the spend being protected. Against it sits a unit that cannot reliably hold negative pressure - which means intermittent loss of AII capability, patient relocation, cancelled admissions, infection prevention investigations, and, if a transmission event occurs, an uninsured clinical and reputational loss. The annual fan energy is roughly the cost of a few days of a blocked isolation unit.
A middle path. I would not simply revert to brute-force constant volume everywhere. A defensible design keeps each room's exhaust at a fixed minimum that guarantees the pressure relationship, uses supply tracking to hold the offset, and lets energy savings come from the parts of the system that do not carry an infection control function - scheduling unoccupied support spaces, duct sealing, coil cleaning, economizer repair. That typically recovers a meaningful share of the $38,000 without touching room pressure stability.
Ask. Approve the controls change and a metered verification period, with the savings estimate re-baselined afterward. I would also flag that the original reprogramming was done without a change-control review by facilities or infection prevention, and recommend that any future setpoint change to a pressure-critical system require that sign-off.