We performed an actigraphic assessment of sleep characteristics in healthy subjects and patients with cognitive impairment. Thirty subjects were included and classified into controls (10 subjects), mild cognitive impairment (10 patients) and mild-to-moderate Alzheimer’s disease (10 patients). Sleep quality was assessed using the Pittsburgh Sleep Quality Index. Participants had a 7-day actigraphic record. Sleep parameters collected were time in bed, total sleep time, sleep efficiency, sleep latency, wakefulness after sleep onset, number of awakenings, and mean motor activity. Significant differences between mild cognitive impairment and controls patients were found for sleep latency (p = 0.05); Alzheimer’s disease patients had significantly worse scores for Pittsburgh Sleep Quality Index (p = 0.01), time in bed (p = 0.001), total sleep time (p = 0.04), sleep latency, sleep efficiency, motor activity (p = 0.0001) and wakefulness after sleep onset (p = 0.001) compared to controls. When comparing Alzheimer’s disease and mild cognitive impairment, differences were significant for sleep latency (p = 0.01), wakefulness after sleep onset (p = 0.004), sleep efficiency, number of awakenings and motor activity (p = 0.0001). In addition to showing a high prevalence of sleep alterations in subjects with cognitive impairment, our data suggest that they are evident from the earliest stages of cognitive decline. Further studies are needed to assess whether early correction of sleep alterations can positively influence the evolution of cognitive impairment. The opportunity to provide clinically meaningful information with a simple assessment of sleep characteristics based on actigraphy suggests that wider use of the approach in patients with cognitive decline should be considered.