These alarms are designed to work by providing warning that the water
surface has been disturbed (e.g., by a toddler falling in), either by detection of
waves on the surface, or motion beneath the surface.
We found two published studies that were laboratory evaluations of pool alarms.
The 1986 study19 demonstrated that in order
for the pool alarm to detect a simulated submersion, the sensitivity of the alarm
had to be set at such a high level as to make the false alarm rate huge (false alarms
from wind, rain, filter start-ups, and unattended use).
A recent study (Whitfield, 2000) evaluated 4 different water disturbance
alarms and wristband alarms; all with remote receivers. The outcomes measured were
detection of surface waves, sub-surface disturbance, exposure to water (for wristband
type alarms), and false alarms. Subsurface pool alarms were the most consistent
and most reliably detected weight of "child like object" entering pool. This type
of alarm produced fewer false alarms than other models.
Pool alarms may supplement, but are not a substitute for supervision or barriers
completely surrounding pools.