AbstractBackgroundSleep disturbances affect 60‐70% of persons living with dementia (PLWD) affecting prognosis and care burden. Sleep studies typically employ actigraphy, however, it is intrusive and lacks information on location. This study explores the use of ambient sensors, to compare nighttime rest patterns of higher and lower functioning PLWD.MethodStudy participants were recruited in a 16‐bed residential care setting for PLWD. Research ethics and consent were obtained. Pressure‐sensing bed mats, motion and door sensors were installed. The sensors were connected to a hub and the cloud.Participants were divided into lower‐ and higher‐functioning groups based on interRAI Home Care scores.An AI program created nightly summaries for each resident using the sensor: evening rest start / end time, number / duration of in room movements (iRM), and actual resting time (aRT) per night.ResultThe higher functioning residents (n = 7) had a mean age of 83.1 ± 11.0 years, as compared to the lower functioning group (n = 5): 85.0 ± 6.7 years. The lower functioning group had consistently higher mean interRAI scores and tended to have a greater number of medications.Preliminary data from 407 nights suggest that the total rest time was longer for the lower functioning group (12.3+/‐0.97 hr) than the higher functioning group (11.9 +/‐ 1.24 hrs). The lower functioning group had 1.83 +/‐ 1.86 iRMs per night, with an average duration of 13.8 +/‐ 11.0 mins. The higher functioning group had 3.34 +/‐ 2.12 iRMs per night, with an average duration of 15.2 +/‐ 9.3 mins. This resulted in an aRT of 11.8 +/‐ 1.04 hrs for the lower functioning group and 11.1 +/‐ 1.25 hrs for the higher functioning group.ConclusionAmbient sensors were used to collect longitudinal data on nighttime activities in PLWD within a communal living setting. The findings suggest evening rest differences between higher‐ and lower‐functioning PLWD. Residents with more advanced dementia appear to be resting more, and those with less impairment seem to have more and longer rest interruptions. This technology could be used to explore the usefulness of various interventions on rest quality.