Accuracy of the GENEActiv Device for Measuring Light Exposure in Sleep and Circadian Research

Abstract

Light is a variable of key interest in circadian rhythms research, commonly measured using wrist-worn sensors. The GENEActiv Original is a cost-effective and practical option for assessing light in ambulatory settings. With increasing research on health and well-being incorporating sleep and circadian factors, the validity of wearable devices for assessing light environments needs to be evaluated. In this study, we tested the accuracy of the GENEActiv Original devices (n = 10) for recording light under a range of ecologically relevant lighting conditions, including LED, fluorescent, infrared, and outdoor lighting. The GENEActiv output had a strong linear relationship with photopic illuminance. However, the devices consistently under-reported photopic illuminance, especially below 100 lux. Accuracy below 100 lux depended on the light source, with lower accuracy and higher variability under fluorescent lighting. The device's accuracy was also tested using light sources of varying spectral composition, which indicated that the device tends to under-report photopic illuminance for green light sources and over-report for red light sources. Furthermore, measures of photopic illuminance were impacted by infrared light exposure. We conclude that the GENEActiv Original is suitable for mapping light patterns within an individual context, and can reasonably differentiate indoor vs. outdoor lighting, though the accuracy is variable at low light conditions. Given the human circadian system's high sensitivity to light levels below 100 lux, if using the GENEActiv Original, we recommend also collecting light source data to better understand the impact on the circadian system, especially where participants spend prolonged periods in dim lighting.

Keywords: actigraphy; ambulatory monitoring; chronobiology; dosimeter; light.

Figure 1

Diagram of device testing configuration. Devices (n = 10) were evenly distributed along a white 56 × 60.5 cm board, with sensors facing upwards towards the light source. The board was 120 cm directly beneath the main light source (or centred between light sources in the case of the spectral performance assessment). Lux meter readings were taken at two points along the devices, marked by blue circles. An infrared camera was located on the ceiling next to the light source. Walls surrounding the board and light source were covered in matte black material. The device configuration on the board was identical for the outdoor light test.

Figure 2

Linearity test of GENEActivs under fluorescent and LED bulbs, and outdoors. (A) Mean GENEActiv lux under fluorescent light (circles), LED light (triangles), and outdoor overcast light (square). Dotted line represents perfect agreement. (B) GENEActiv lux (n = 10) relative (%) to the lux meter under fluorescent light; each symbol represents one device. (C) GENEActiv lux (n = 10) relative (%) to the lux meter under LED light; each symbol represents one device.

Figure 3

GENEActiv output under ~120 photopic lux with varying wavelength: red (630 nm), blue (470 nm), and green (520 nm), with error bars (standard deviation from the mean). Diamond indicates GENEActiv output under infrared light at 0 lux (note the error bars for infrared are very small and therefore difficult to see; see Table 3). The dotted line indicates perfect agreement between the lux meter and GENEActiv output.