The relationship of light exposure to sleep outcomes among office workers. Part 1: Working in the office versus at home before and during the COVID-pandemic

Abstract

The relationship between everyday light exposure and sleep was studied for office workers. The study was conducted during the upswing of the COVID-19 pandemic, enabling a comparison between Office and Home Workdays. Fifteen full-time office employees were monitored for a period of 4–6 weeks. They wore a light-tracking device on their clothes and had a sleep tracker at home. Compared to an Office Workday, light exposure was lower in the afternoon and total sleep time was almost 5 minutes longer on a Home Workday. Sleep efficiency was the same on both workday types. A higher median illuminance level in the afternoon was significantly related to later sleep onset on an Office Workday. Higher median illuminance levels in the morning were related to earlier awakening. Counter to expectations, higher light levels in the evening were also related to earlier awakening. Everyday light exposure matters for sleep quality but may affect circadian functioning differently than the often more extreme light interventions employed in laboratory experiments. Moreover, differences in outcomes between Office and Home Workdays signal the need for further investigation to provide supportive light levels during workhours.

1. Introduction

A good night’s sleep gives the best start to the day. Sleep has been linked to mental and physical health aspects, ranging from better mood to lower overall mortality.^1–3 Sleep is negatively affected by workload, work hours and perseverative cognition^4,5 and personal factors such as having small children or not. Depending on its level, spectral distribution, timing and duration, exposure to light can affect human sleep and circadian functioning.^6,7 In modern society, it is no longer (only) the sun that dictates the human rest-activity cycles.^8,9 Social constraints (i.e. work schedules) and electric lighting available at any time affect sleep-wake rhythms in parallel. With an 8-hour working day, office workers spend about half of their time awake in office environments and, additionally, before and after work, considerable time at home. This activity pattern makes office and home lighting significant contributors to daily light exposure.¹⁰ Light levels in indoor environments are significantly lower (often <1000 lx) than exposure to daylight outdoors (often >1000 lx).^7,11,12 Still, it is not only bright light exposure over 1000 lx that affects circadian functioning.^12,13 Low to medium light levels have been reported to affect well-being and sleep.^6,14–16 Especially in a Scandinavian setting, lighting at home usually has lower correlated colour temperatures than lighting used in commercial buildings. Traditionally this means that the lighting had a lower short-wavelength content. However, Cain et al.¹⁷ noted that homes with energy-efficient lamps had higher evening light exposure and consequently poorer sleep efficiency, potentially as energy-efficient fluorescent and LED lighting tends to double the amount of melanopic daylight-equivalent illuminance (mEDI) levels compared to incandescent light sources.

Light effects on sleep have been investigated in naturalistic, ambulatory studies looking at human light exposure and (self-reported) sleep quality, usually for 7 consecutive days.^18–22 These studies looked at all-day light exposure, all-day short-wavelength spectral irradiance, morning/mid-day/evening exposure, and the proportion of bright light exposure, sometimes in different seasons. The studies found significant effects of daytime exposure on sleep, including earlier mid-sleep after higher levels of morning light exposure²⁰ and more pronounced effects in winter,¹⁸ but not significant relations between mid-day exposure and sleep.²⁰ Nevertheless, in some cases, light exposure was measured via a wearable worn on the wrist²⁰ or upper arm.²² A recent systematic review on daily light exposure, sleep-wake rhythm, and mood reported limited and conflicting evidence for a positive relationship, largely due to the lack of intervention studies and the use of wrist worn light exposure devices.²³

In the spring of 2020, the COVID-pandemic made several governments and public health authorities worldwide order or recommend employees to work from home, if possible. Office workers typically spend a substantial part of their waking hours in the office. This shift to working from home affected many aspects of the work environment, including light exposure. Lighting environments may differ between the office and the home, as lights in the home environments are usually more focused on creating a cosy atmosphere.²⁴ Working from home may also induce behavioural shifts that can impact light exposure. First, workers may no longer need to commute from their home to work. This outdoor commute usually contributes considerably to the daily light exposure.^12,25 In addition, with disappearing commute time, people tend to change their sleeping times.²⁶ Third, a recent study has indicated that office workers spent more time behind a screen during the COVID-19 pandemic.²⁷ Screen time can significantly contribute to light exposure; particularly, increased screen time in the evening can delay sleep onset.^28,29 Xiao et al.³⁰ showed that, during Home Workdays, overall mental and physical well-being decreased due to, for example, the lack of physical exercise or communication with co-workers, changed food intake, adjusted work hours, or the workstation set-up. In addition to social and behavioural changes, prolonged sedentary activity due to continuous online meetings at improvised desks (i.e. kitchen table) lacking proper electric lighting and daylight access can undoubtedly affect an employee’s light exposure.

The effects of light exposure on sleep have been proposed to depend not only on the light amount, spectral composition and timing of the light but also on, for example, season,³¹ duration,³² or light history.³³ Therefore, in the current study, office worker’s naturally occurring light exposure and sleep were monitored for multiple consecutive weeks, both during and outside working hours, using a 24-hour ambulatory assessment during one season (spring). The present study started when the upswing of the COVID-19 pandemic was lurking, but no measures were announced or instituted yet. During the study, COVID-measures were implemented, and employees were asked to work from home as much as possible.

The present study captured potential differences in sleep quality/patterns and light exposure between working from home and working in the office. The effects of light exposure on sleep outcomes were investigated at three times: morning, afternoon and evening. Light variables under study were (vertical) illuminance level and red (R), green (G) and blue (B) values. Due to the high unreliability of the RGB values, these results were not considered. It was hypothesised that:

An increase in light exposure in the morning results in higher sleep efficiency and earlier sleep onset and offset,

An increase in light exposure in the evening results in lower sleep efficiency and later sleep onset and offset (i.e. an opposite, but less-pronounced pattern from the morning exposure),

An increase in afternoon light exposure does not influence sleep outcomes since it is outside the critical period for sleep advance (between 06.00 and noon) and sleep delay (midnight and 06.00), and

Light exposure was lower on Home Workdays than on Office Workdays, which would negatively affect sleep outcomes.

2. Method

In an ambulatory field study, 15 participants worked from home more often during the second half of the study, which monitored their personal light exposure and sleep quality for 4–6 weeks.

2.1 Participants

Office employees of two different companies and within different buildings in Sweden (~57°N, 14°E) were invited to participate. In total, 15 participants (seven females, age between 29 and 63 years, M = 43.3 ± 10.4) were included in this study. If a participant was pregnant, used sleep medication, was professionally diagnosed with a sleep disorder, worked less than 4 days per week, or had children younger than 1 year, he/she was excluded from participation. An online questionnaire was administered before and after the experiment. It included questions on demographics (age, gender, weight, height, household arrangement), commuting behaviour, regular working hours, sports and leisure activities, and long-term subjective sleep quality (Pittsburgh Sleep Quality Index³⁴).

2.2 Materials

Personal light exposure was measured with a wearable light detector (LYS Button by LYS Technologies LTD™) at a 15-second interval. The light detector was worn as close to the eyes as possible during the daytime and placed on the bedside table at night. The detector measures illuminance (in lux) as well as (R), (G) and (B) 16-bit values. Validation is performed on request of the manufacturer itself, but the results are not published. All 15 light detectors were checked and calibrated prior to the study, both for indoor and outdoor light conditions, against two Hagner EC1 illuminance meters (calibrated by B. Hagner AB in February 2019 with limited use after; calibration accuracy ±3%). A correction factor was applied to all detectors. Differences for indoor conditions were between 0 and 5%, while outdoor values showed much higher differences. Thus, for typical indoor conditions (up to 5000 lx), the light detector performs appropriately. Additionally, the light trackers were investigated for their performance regarding spectral ranges for ‘R’, ‘G’ and ‘B’. However, the sensitivity curves for the B and G channels had a significant overlap, and hence the RGB values were not further considered in the study. Sleep variables were measured with a non-contact sleep tracking device (SleepScore Max by SleepScore Labs™) based on sonar and radiofrequency. The device analyses sleep quality entirely without contact with the participant as it uses bio-motion sensor technology, like echolocation, to track breathing and body movement. Validation is performed on request of the manufacturer itself by independent institutes and the results are published in peer-reviewed scientific journals.^35,36 For example, Schade et al.³⁶ compared the device against polysomnography and showed that the devices typically had a high sleep-detecting sensitivity (⩾95%) but poor wake-detecting specificity (⩽40%).

All participants received a 4G-equipped mobile phone (iPhone 6) with a charger containing data collection apps. The light tracker used the regular app (without data reveal as it went via a paid service). The sleep tracker used a special research edition of the app to prevent data reveal to the participants (‘R’ version).

The study was executed in two offices with open-plan and single office spaces in Fagerhult and Jönköping, Sweden. Building one is equipped with suspended direct/indirect luminaires (Fagerhult Notor 78; 50/50 or Fagerhult Notor Skywalker, 40/60) above each desk. The other building has ceiling-mounted direct luminaires (Fagerhult Multilume Slim) in the open-plan offices and suspended direct/indirect luminaires (Fagherhult Itza Delta, 50/50) in the single rooms. The lighting of both office buildings was connected to presence sensors. Daylight can enter the room via the windows unless this is prevented using Venetian blinds or screens. The daylight contribution in the office per orientation was logged every 5 minutes with an illuminance logger (YoYo 2YL-M61-4M/2YL-M62-4M by Grant Instruments) horizontally placed at the windowsill (nine loggers total). The vertical illuminance at each desk was logged using an illuminance logger (HOBO MX2202 by Onset Computer Corporation), hanging at the desk divider at eye height. Presence at the desk was registered using a wireless occupancy desk sensor and logger (iotspot Hub).

2.3 Procedure

The data collection took place both in the office and at home. Participants were followed for at least 4 weeks during late winter (between February 27th and April 6th). Prior to the experiment, for both companies separately, the experiment’s general set-up and procedure were explained. The participants were aware of partaking in an experiment related to lighting and sleep quality but unaware of the exact study purpose. All participants got a unique ID code to ensure confidentiality of their identity and gave their written consent for participation and data collection. The study protocol was approved by the Swedish Ethical Review Authority (no. 2019-05453).

2.4 Data analysis

Given the nested nature of the data, hierarchical linear models were used to test relationships between light exposure and sleep outcomes, with responses nested within individuals as well as within days. Separate models were run for the morning (06.00–11.00 hours), afternoon (12.00–17.00 hours) and evening (18.00–23.00 hours). Three sleep variables were included as outcome variables: sleep efficiency (ratio of wake time after sleep onset and total sleep time), sleep onset and sleep offset. Three different models were run: one Overall model looking at all workdays together and two models looking at Home Workdays and Office Workdays separately.

Median illuminance level per hour was added as a predictor. Data from the occupancy sensors were used to determine whether participants were in the office (>10% presence) or not (<10% presence). In this data, no distinction could be made between working at home, working outside the office the entire day (e.g. visiting clients even though unlikely during the pandemic), and days off on a weekday.

Only hours for which at least 40% of the data was above a threshold of 10 lx were included. As the illuminance values ranged between very low and very high peak values (e.g. between 10 and 29,825 lx), all values were log-transformed. Within the study period, the clock was set to summertime (March 29th). Therefore, data from 3 days after daylight saving time were deleted from the analyses.

3. Results

Several indicators for sleep and light exposure were monitored for office employees for at least 4 weeks during and after work time. As participants were recruited from two different companies, before starting the main analyses, it was tested whether outcome variables or light exposure differed between the two participant groups. They did not differ on any of the variables (all p > 0.155).

A total of 5550 valid hourly light measurements were included in the analyses, ranging from 90 to 452 measurements per participant, with a mean of 330 measurements. Missing data points could be due to user behaviour (e.g. going outdoors with a coat over the light sensor or forgetting to synchronise the device) or technical issues (e.g. no connection). As the participants had to switch on the sleep tracking device each night manually, there were some missing sleep data. A total of 355 sleep episodes were recorded, ranging between 13 and 30 per participant, with a mean of 24 measurements.

The percentage Office Workdays per person ranged from 20% to 88%, with an average of 0.49% (SD = 0.28). The percentage of Office Workdays was highest during the second week of the study, and lowest during the last week of the study (week 1: 64%, week 2: 68%, week 3: 45%, week 4: 36%, week 5: 33%).

After the study was finished, three questions were asked to measure how the COVID-19 pandemic had influenced the participants. These questions concerned to what degree they felt informed about COVID-19, whether COVID-19 had impacted their lives, and how much they ruminated about COVID-19, see Figure 1 for the answers. No significant relation was found between these COVID-19 related questions and any sleep outcomes (all p > 0.293). In addition, the correlation between how much people worried about COVID-19 and the number of home working days was nonsignificant (p = 0.687).

Figure 1

Outcomes of the COVID-19 pandemic control questions: being informed (a), impacted (b) and worried (c)

3.1 Sleep outcomes

Data were recorded for 195 workdays ‘at the office (Office Workday)’ and 108 workdays ‘not at the office (Home Workday)’. The data recorded for 105 ‘Weekend’ days were used in Beute et al.³⁷ Sleep parameters on a Home Workday were compared to Office Workdays; see Table 1 for an overview of sleep outcomes for each type of day. Compared to an Office Workday, on Home Workdays, sleep onset was 6 minutes later, midpoint sleep was 9 minutes later, and sleep offset was 16 minutes later. Total sleep time was approximately 5 minutes longer on a Home Workday than an Office Workday and sleep efficiency was the same on Office Workdays and Home Workdays. See Figure 2 for an overview of the sleep times.

Table 1

Overview of the estimated marginal means (EMMs) with standard errors (SEs) for sleep outcomes for an Office Workday and a Home Workday

	EMM (SE) sleep outcomes
	Office Workday	Home Workday
Sleep onset	23.20 (10.3 minutes)	23.26 (10.4 minutes)*
Sleep offset	06.46 (10.9 minutes)	07.02 (10.9 minutes)*
Midpoint sleep	02.57 (9.5 minutes)	03.06 (9.5 minutes)*
Total sleep time	395 minutes (11.36 minutes)	400 minutes (11.41 minutes)*
Sleep efficiency	89.44% (1.55)	89.80% (1.55)

p < 0.05.

Figure 2

Sleep onset, midpoint sleep (circle), and sleep offset on Office Workdays and Home Workdays

3.2 Personal light exposure

Light exposure differed between the two different workdays; see Table 2. Compared to an Office Workday, light exposure was lower in the afternoon on a Home Workday. Figure 3 displays the median light exposure across the day (median log-transformed illuminance). Afternoon light exposure was positively skewed in terms of light level, and results for the afternoon need to be interpreted with caution.

Table 2

Overview of EMMs with SEs for illuminance levels (log-transformed) for the 2 workday types for the three different hierarchical linear models (HLM) models

HLM model	EMM (SE) illuminance levels
HLM model	Office Workday	Home Workday
Morning	2.19 (0.03)	2.14 (0.04)
Afternoon	2.23 (0.04)*	2.10 (0.04)*
Evening	1.48 (0.04)	1.49 (0.05)

p < 0.05.

Figure 3

Mean values for the log-transformed median illuminance levels across an Office Workday and a Home Workday. The hours between 23.00–5.00 were excluded due to a low number of observations per type of day (< 20)

3.3 Relation light exposure and sleep outcomes

Table 3 shows the outcomes of the three different models for median illuminance level. Sleep offset on a subsequent day was earlier with a higher median illuminance level in the morning for both the Overall model and the Office Workday model. No effect of morning median light level on sleep offset was found for the Home Workday. In the afternoon, a higher median illuminance level was related to later sleep onset, but only for the Office Workday model. In the evening, the median illuminance level was negatively associated with sleep offset for the Overall model, with later sleep offset after being exposed to higher evening light levels. Additionally, median illuminance level was significantly related with sleep offset for an Office Workday, again with later sleep offset after being exposed to higher light levels. On a Home Workday, this relation did not reach significance.

Table 3

The unstandardised estimates (EST) with SEs, 95% confidence intervals (CIs) (lower (LCI) and upper confidence interval (UCI)), and intra class correlations (ICC) of the sleep outcomes for the three models for light exposure measured by the median illuminance level

	Sleep efficiency				Sleep onset				Sleep offset
	EST (SE)	LCI	UCI	ICC	EST (SE)	LCI	UCI	ICC	EST (SE)	LCI	UCI	ICC
Morning
Overall	0.54 (0.37)	−0.18	1.26	0.64	−0.96 (2.90)	−6.65	4.73	0.59	−6.01 (1.78)**	−9.50	−2.52	0.85
Office	0.81 (0.49)	−0.16	1.78	0.63	0.26 (3.17)	−5.95	6.47	0.65	−4.20 (2.04)*	−8.20	−0.20	0.87
Home	0.32 (0.45)	−0.56	1.20	0.78	1.50 (2.29)	−2.99	6.00	0.97	−2.82 (2.31)	−7.35	1.70	0.95
Afternoon
Overall	0.24 (0.41)	−0.55	1.04	0.58	6.02 (3.17)	−0.19	12.23	0.55	−1.14 (1.80)	−4.68	2.93	0.85
Office	0.66 (0.55)	−0.40	1.74	0.57	14.96 (3.37)**	8.36	21.56	0.59	1.05 (2.01)	−2.90	5.00	0.88
Home	−0.23 (0.50)	−1.21	0.74	0.77	−0.11 (2.14)	−4.31	4.09	0.97	1.24 (2.23)	−3.14	5.61	0.94
Evening
Overall	−0.05 (0.63)	−1.29	1.19	0.54	−9.17 (4.90)	−18.77	0.42	0.54	−7.82 (2.55)*	−12.82	−2.82	0.86
Office	−0.33 (0.82)	−1.93	1.26	0.54	−6.95 (4.37)	−15.51	1.51	0.63	−6.26 (2.88)*	−11.91	−0.62	0.87
Home	0.53 (0.76)	−0.94	2.02	0.70	−3.72 (3.35)	−10.30	2.85	0.97	−1.74 (2.09)	−5.84	2.37	0.98

p < 0.05. **p < 0.001.

4. Discussion

In an ambulatory field study, office workers’ light exposure and sleep quality data were collected. The study captured personal light exposure from waking up in the morning to going to sleep at night during the start of the COVID-19 pandemic.

Due to the COVID-19 pandemic, participants worked from home more than usual as the study progressed. In Sweden, there was no strict lockdown, and therefore, social constraints on sleep times during Home Workdays might differ less from Office Workdays than in other countries with (stricter) lockdowns.³⁸ In addition, not all participants were requested to work from home daily. Still, participants got out of bed, on average, 16 minutes later on Office Workdays than on Home Workdays and went to bed, on average, 6 minutes later. They also slept around 5 minutes longer, which is similar to what is reported in countries with a lockdown.³⁹ No differences in sleep efficiency were found between Office and Home Workdays.

Three different models investigated the relationship between light exposure and sleep outcomes. Previous studies have indicated that home lighting often has lower intensities than office lighting.^19,40 Nevertheless, results of a 3-week study investigating evening light levels demonstrated that self-chosen home light levels before habitual bedtime can increase circadian misalignment.⁴¹ even though there is often a wide range of individual responses.¹⁷ In the current field study, only lower afternoon light exposure was reported on a Home Workday compared to an Office Workday. Differences in light exposure might have been more pronounced if the effects of the lockdown had been studied during the mid-winter season.

A recent overview by Vetter et al.⁴² showed that bright light exposure in the morning phase advances sleep on a consecutive night and bright light exposure in the second part of the biological day phase delays sleep on a consecutive night. The outcomes of the present study only partly followed this expected pattern. In line with expectations, higher levels of light in the morning were related to an earlier offset (waking up) the day after for the Overall and the Office Workday model. However, counter to expectations, higher evening light exposure was related to an earlier offset in the Overall and Office Workday models. No significant relationships were found between morning and evening light exposure at a Home Workday and sleep outcomes. Potentially, these differences are due to lower statistical power as there were fewer Home Workdays in the sample than Office Workdays even though the model coefficients did point in the same direction.

The counterintuitive outcomes for the evening may be due to several factors. Previous research has pointed at a large variability in sensitivity to evening exposure^14,17 and the importance of light history^33,43–45; for instance, exposure to blue-enriched light in the morning or bright daytime light exposure counteracting higher light levels in the evening. In other words, effects of evening exposure may depend on how much light a person received in the morning (and when and with which composition). Phase-delaying effects of bright or blue-enriched electric light in the evening are often found in laboratory studies. Laboratory studies may differ in timing (with night-time light exposure as opposed to evening light exposure), duration (longer duration to higher light levels), and intensity/composition (using, for example, very bright light exposure and blue-enriched light). The test conditions may differ from the naturally occurring light levels in the present study. Potentially, the differences in light exposure in terms of vertical illuminance level in the present study are of a much more subtle nature than in laboratory experiments and therefore resulted in different outcomes. Alternatively, differences in outcomes compared to earlier studies could be due to the extended investigation time (4–6 weeks rather than 7 days). Detrimental effects of light exposure mostly appear later when participants had often already gone to bed on workdays. In addition, the results show that on both workday types, there was an increase in illuminance level after 20.00. Previous research has found that participants usually spend the last hour before going to bed watching television.⁴⁰ An increased exposure might be explained by people continuing to work in the evening (with increased levels of electric lighting). Additionally, research has indicated that people spend more time behind a screen during the COVID-19 pandemic.²⁷

No significant relationship was found with the sleep outcomes on the Home Workday or the Overall model for afternoon light exposure. However, for the Office Workday, a higher median illuminance level was found related to a later sleep onset. This effect only occurred for the Office Workday. As a significantly higher light exposure on the Office Workday compared to the Home Workday was found, and the direction of the (nonsignificant) relationship between afternoon light exposure and sleep outcomes for the other two models were in the opposite direction, it seems plausible that these variations were due to differences in light environments. However, as light exposure was positively skewed in the afternoon, this result needs to be corroborated in future research. Research often focuses on morning and evening light exposure; therefore, less is known about the effects of afternoon exposure to light on sleep. Other studies have not found a relationship between an afternoon or mid-day light exposure and sleep outcomes.^20,46,47 The absence of a significant relationship between sleep quality/duration and light exposure in the morning or afternoon can possibly be due to the use of subjective sleep assessment tools (i.e. sleep diary⁴⁸) rather than an objective instrument.

The applied light detector offered RGB-values measurement, which could have given – if sufficiently distinctive – an indication regarding the light exposure being,that is more short-wavelength enriched or deprived, even though a previous study reported a high correlation between RGB clusters and illuminance level.¹⁹ Nevertheless, daylight varies in spectral power distribution over the day, especially when comparing twilight versus midday compositions. This calls for studying more subtle daily variations in spectral power composition and their relationship with everyday light effects and sleep outcomes. Especially since it is not only the short wavelengths of light that deserve attention but also other wavelengths of the visual spectrum in daily life. At present, only a few studies investigated effects of longer wavelengths of light, often only as a control or placebo condition.

Several factors pose limitations to the present study. First, technical failures of the light detector and omissions of the participants switching on the sleep device led to missing data. Furthermore, as participants were instructed to wear the light tracker on their clothes, outdoor light exposure was not (correctly) captured when a coat blocked the sensor. Third, the study was not explicitly designed to capture differences between home and Office Workdays but caught them inadvertently. Differences in daily routines (e.g. not traveling to work) may have affected light exposure. In addition, even though participants occasionally worked from home in the beginning of the study, the frequency was higher during the end of the study. Small differences in light exposure due to changes in daylight hours may have affected the outcomes. Altered daily activities and mood changes or rumination levels due to the COVID-19 pandemic may have affected sleep during the experiment, besides light exposure. Last, age differences between participants may have led to differences in sleep outcomes and circadian effects of light, as older adults do not sleep as well as younger adults. For example, older participants (57–74 years) in the study of Münch et al.⁴⁹ felt significantly sleepier and reported more sleep at circadian times corresponding to the late afternoon and evening (16.00–22.00 hours) compared to younger ones (20–31 years). As a result of ageing, some older people have difficulties staying awake during evening hours due to a weakening circadian wake signal. Fortunately, this is not the case for teenagers and young adults.⁵⁰

The present study found that everyday light environments matter for sleep outcomes. In addition, differences between office light environments and home light environments may influence how employees sleep on an Office versus Home Workday but needs further investigation especially for the afternoon. Everyday light exposure appeared to affect sleep differently from results found in laboratory, especially with regards to evening light exposure. Potentially, the more subtle daily fluctuations in light exposure in everyday life, as opposed to the more extreme light manipulations in laboratory studies, may result in different relationships between light exposure and sleep.

Footnotes

Acknowledgements

The authors would like to thank the participants and office managers in the field study offices for their input, feedback, and efforts in facilitating this project. The free-of-charge provision of the SleepScore Max sleep analysers, including a special research application by SleepScore Labs™ is particularly appreciated.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Bertil & Britt Svenssons Stiftelse för Belysningsteknik [2018-11-26] and the Department of Construction Engineering and Lighting Science’s Internal Strategic Funds [2019-01-30].

ORCID iDs

A Lowden

F Beute

References

Konjarski

Murray

Lee

Jackson

ML.

Reciprocal relationships between daily sleep and mood: a systematic review of naturalistic prospective studies. Sleep Medicine Reviews 2018; 42: 47–58.

Gallicchio

Kalesan

Sleep duration and mortality: a systematic review and meta-analysis. Journal of Sleep Research 2009; 18: 148–158.

Blume

Garbazza

Spitschan

Effects of light on human circadian rhythms, sleep and mood. Somnologie (Berl) 2019; 23: 147–156.

Akerstedt

Fredlund

Gillberg

Jansson

Work load and work hours in relation to disturbed sleep and fatigue in a large representative sample. Journal of Psychosomatic Research 2002; 53: 585–588.

Clancy

Prestwich

Caperon

Tsipa

O’Connor

DB.

The association between worry and rumination with sleep in non-clinical populations: a systematic review and meta-analysis. Health Psychology Review 2020; 14: 427–448.

Tahkamo

Partonen

Pesonen

AK.

Systematic review of light exposure impact on human circadian rhythm. Chronobiology International 2019; 36: 151–170.

Wirz-Justice

Skene

Münch

The relevance of daylight for humans. Biochemical Pharmacology 2021; 191: 114304.

Wright

Jr McHill

Birks

Griffin

Rusterholz

Chinoy

. Entrainment of the human circadian clock to the natural light-dark cycle. Current Biology 2013; 23: 1554–1558.

Wittmann

Dinich

Merrow

Roenneberg

Social jetlag: misalignment of biological and social time. Chronobiology International 2006; 23: 497–509.

10.

Heil

Mathis

SR.

Characterizing free-living light exposure using a wrist-worn light monitor. Applied Ergonomics 2002; 33: 357–363.

11.

Aries

MBC

Aarts

MPJ

van Hoof

. Daylight and health: a review of the evidence and consequences for the built environment. Lighting Research and Technology 2015; 47: 6–27.

12.

Dumont

Beaulieu

Light exposure in the natural environment: relevance to mood and sleep disorders. Sleep Medicine 2007; 8: 557–565.

13.

Burns

Saxena

Vetter

Phillips

AJK

Lane

Cain

SW.

Time spent in outdoor light is associated with mood, sleep, and circadian rhythm-related outcomes: a cross-sectional and longitudinal study in over 400,000 UK Biobank participants. Journal of Affective Disorders 2021; 295: 347–352.

14.

Phillips

AJK

Vidafar

Burns

McGlashan

Anderson

Rajaratnam

SMW

, et al. High sensitivity and interindividual variability in the response of the human circadian system to evening light. Proceedings of the National Academy of Sciences (PNAS) 2019; 116: 12019–12024.

15.

Gooley

Chamberlain

Smith

Khalsa

SBS

Rajaratnam

Van Reen

, et al. Exposure to room light before bedtime suppresses melatonin onset and shortens melatonin duration in humans. Journal of Clinical Endocrinology and Metabolism 2011; 96: E463–E472.

16.

Aries

MBC

Beute

Fischl

. Assessment protocol and effects of two dynamic light patterns on human well-being and performance in a simulated and operational office environment. Journal of Environmental Psychology 2020; 69: 101409.

17.

Cain

McGlashan

Vidafar

Mustafovska

Curran

SPN

Wang

, et al. Evening home lighting adversely impacts the circadian system and sleep. Scientific Reports 2020; 10: 19110.

18.

Figueiro

Steverson

Heerwagen

Kampschroer

Hunter

Gonzales

, et al. The impact of daytime light exposures on sleep and mood in office workers. Sleep Health 2017; 3: 204–215.

19.

Hubalek

Brink

Schierz

Office workers’ daily exposure to light and its influence on sleep quality and mood. Lighting Research and Technology 2010; 42: 33–50.

20.

Lowden

Lemos

NAM

Goncalves

BSB

Ozturk

Louzada

Pedrazzoli

, et al. Delayed sleep in winter related to natural daylight exposure among arctic day workers. Clocks and Sleep 2019; 1: 105–116.

21.

Shochat

Santhi

Herer

Flavell

Skeldon

Dijk

DJ.

Sleep timing in late autumn and late spring associates with light exposure rather than sun time in college students. Frontiers in Neuroscience 2019; 13: 882.

22.

Amdisen

Daugaard

Vestergaard

Vested

Bonde

Vistisen

, et al. A longitudinal study of morning, evening, and night light intensities and nocturnal sleep quality in a working population. Chronobiology International 2022; 39: 579–589.

23.

Bohmer

Hamers

PCM

Bindels

PJE

Oppewal

van Someren

EJW

Festen

DAM

. Are we still in the dark? A systematic review on personal daily light exposure, sleep-wake rhythm, and mood in healthy adults from the general population. Sleep Health 2021; 7: 610–630.

24.

Gerhardsson

Laike

Johansson

Leaving lights on –a conscious choice or wasted light? Use of indoor lighting in Swedish homes. Indoor and Built Environment 2021; 30: 745–762.

25.

van Duijnhoven

Aarts

Aries

MBC

Bohmer

Rosemann

AL.

Recommendations for measuring non-image-forming effects of light: a practical method to apply on cognitive impaired and unaffected participants. Technology and Health Care 2017; 25: 171–186.

26.

Korman

Tkachev

Reis

Komada

Kitamura

Gubin

, et al. COVID-19-mandated social restrictions unveil the impact of social time pressure on sleep and body clock. Scientific Reports 2020; 10: 22225.

27.

Majumdar

Biswas

Sahu

COVID-19 pandemic and lockdown: cause of sleep disruption, depression, somatic pain, and increased screen exposure of office workers and students of India. Chronobiology International 2020; 37: 1191–1200.

28.

Chang

Aeschbach

Duffy

Czeisler

CA.

Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. Proceedings of the National Academy of Sciences 2015; 112: 1232–1237.

29.

Arend

Blechert

Pannicke

Reichenberger

Increased screen use on days with increased perceived COVID-19-related confinements - a day level ecological momentary assessment study. Front Public Health 2020; 8: 623205.

30.

Xiao

Becerik-Gerber

Lucas

Roll

SC.

Impacts of working from home during COVID-19 pandemic on physical and mental well-being of office workstation users. Journal of Occupational and Environmental Medicine 2021; 63: 181–190.

31.

Adamsson

Laike

Morita

Seasonal variation in bright daylight exposure, mood and behavior among a group of office workers in Sweden. Journal of Circadian Rhythms 2018; 16: 2.

32.

Martinez-Nicolas

Ortiz-Tudela

Madrid

Rol

MA.

Crosstalk between environmental light and internal time in humans. Chronobiology International 2011; 28: 617–629.

33.

Chang

Scheer

Czeisler

CA.

The human circadian system adapts to prior photic history. Journal of Physiology 2011; 589: 1095–1102.

34.

Buysse

Reynolds

3rd Monk

Berman

Kupfer

DJ.

The pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Research 1989; 28: 193–213.

35.

O’Hare

Flanagan

Penzel

Garcia

Frohberg

Heneghan

A comparison of radio-frequency biomotion sensors and actigraphy versus polysomnography for the assessment of sleep in normal subjects. Sleep and Breathing 2015; 19: 91–98.

36.

Schade

Bauer

Murray

Gahan

Doheny

Kilroy

, et al. Sleep validity of a non-contact bedside movement and respiration-sensing device. Sleep 2017; 40: 1051–1061.

37.

Beute

Lowden

Aries

MBC

. The relationship of light exposure to sleep outcomes among office workers. Part 2: Comparison of days with and without social constraints. Lighting Research and Technology 2024; 56: 126–135.

38.

Casagrande

Forte

Tambelli

Favieri

The coronavirus pandemic: a possible model of the direct and indirect impact of the pandemic on sleep quality in Italians. Nature and Science of Sleep 2021; 13: 191–199.

39.

Blume

Schmidt

Cajochen

Effects of the COVID-19 lockdown on human sleep and rest-activity rhythms. Current Biology 2020; 30: R795–R797.

40.

Crowley

Molina

Burgess

HJ.

A week in the life of full-time office workers: work day and weekend light exposure in summer and winter. Applied Ergonomics 2015; 46: 193–200.

41.

Burgess

Molina

TA.

Home lighting before usual bedtime impacts circadian timing: a field study. Photochemistry and Photobiology 2014; 90: 723–726.

42.

Vetter

Pattison

Houser

Herf

Phillips

AJK

Wright

, et al. A review of human physiological responses to light: implications for the development of integrative lighting solutions. LEUKOS 2021; 18: 387–414.

43.

Rangtell

Ekstrand

Rapp

Lagermalm

Liethof

Bucaro

, et al. Two hours of evening reading on a self-luminous tablet vs. reading a physical book does not alter sleep after daytime bright light exposure. Sleep Medicine 2016; 23: 111–118.

44.

Porcheret

Wald

Fritschi

Gerkema

Gordijn

Merrrow

, et al. Chronotype and environmental light exposure in a student population. Chronobiology International 2018; 35: 1365–1374.

45.

Pauley

SM.

Lighting for the human circadian clock: recent research indicates that lighting has become a public health issue. Medical Hypotheses 2004; 63: 588–596.

46.

Gimenez

Hessels

van de

Werken M

de Vries

Beersma

Gordijn

MC.

Effects of artificial dawn on subjective ratings of sleep inertia and dim light melatonin onset. Chronobiology International 2010; 27: 1219–1241.

47.

Carrier

Dumont

Sleep propensity and sleep architecture after bright light exposure at three different times of day. Journal of Sleep Research 1995; 4: 202–211.

48.

Peeters

Smolders

KCHJ

Vogels

IMLC

de Kort

YAW

. Less is more? Effects of more vs. less electric light on alertness, mood, sleep and appraisals of light in an operational office. Journal of Environmental Psychology 2021; 74: 101583.

49.

Münch

Knoblauch

Blatter

Schroder

Schnitzler

Krauchi

, et al. Age-related attenuation of the evening circadian arousal signal in humans. Neurobiology of Aging 2005; 26: 1307–1319.

50.

Cajochen

Garbazza

. Circadian biology/chronobiology - physiological basis of sleep. In Amici

Peigneux

, editors. Sleep Medicine Textbook 2021. Regensburg, Germany: European Sleep Research Society (ESRS), 2021: 123–134.