Can Avoiding Light at Night Reduce the Risk of Breast Cancer?

Abstract

Excessive exposure to artificial light at night (ALAN) suppresses nocturnal melatonin (MLT) production in the pineal gland and is, therefore, associated with an increased risk of breast cancer (BC). We examined indoor and outdoor light habits of 278 women, BC patients (n = 93), and controls (n = 185; 2010-2014). Cases and controls were age and residential area matched. Data regarding behavior in the sleeping habitat in a 5-year period, 10 to 15 years prior to disease diagnosis, were collected using a questionnaire. Sleep quality, bedtime, sleep duration, TV watching habits, presleeping reading habits, subjective illumination intensity, and type of illumination were collected. Binary logistic regression models were used to calculate odds ratios with 95% confidence intervals (ORs with 95% CIs) for BC patients in relation to those habits. OR results revealed that women who had slept longer (controls), 10 to 15 years before the time of the study, in a period of 5 years, had a significant (OR = 0.74; 95% CI = 0.57-0.97; P < .03) reduced BC risk. Likewise, women who had been moderately exposed to ALAN as a result of reading using bed light (reading lamp) illumination and women who had slept with closed shutters reduced their BC risk: OR = 0.81, 95% CI = 0.67-0.97, P < .02, and OR = 0.82, 95% CI = 0.68-0.99, P < .04, respectively. However, women who had been exposed to ALAN as a result of living near strong illumination sources were at a significantly higher BC risk (OR = 1.52; 95% CI = 1.10-2.12; P < .01). These data support the hypothesis that diminishing nighttime light exposure will diminish BC risk and incidence. This hypothesis needs to be tested directly using available testing strategies and technologies that continuously measure an individual’s light exposure, its timing, and sleep length longitudinally and feed this information back to the individual, so that BC risk can be distinguished prospectively.

Keywords

artificial light at night breast cancer latency period short wavelength illumination urban rural

Introduction

Daily, lunar, and seasonal cycles of natural light are environmentally temporal cues for all mammals, including humans.^1,2 In the past 100 years, natural light cycles have been greatly disrupted by the introduction of artificial light at night (ALAN) emerging from a diversity of sources, including street and bill board lightings, decorative and domestic lightings, security lighting, and lighting originating from transportation.^3-5 Natural light/dark cycles are the major zeitgeber for the entrainment of the biological clock, enabling the coordination of body physiology, immune and behavioral functions with daily and seasonal changes.⁶ Therefore, exposure to ALAN may affect humans’ physiology, immune function, and health.⁷ As a consequence of its effect on human health, in recent years, exposure to ALAN has become known as a source of possible circadian disruption for individuals exposed to the same in places, times, and intensities in which natural light does not occur.

As natural light changes in its intensity and wavelength throughout the day, it entrains our biological clock. In contrast to natural light, ALAN contains light with different spectra and intensities, and therefore, affects natural daily rhythms.^1,4 The biological clock is a mechanism consisting of a central oscillator located in the hypothalamus region (suprachiasmatic nucleus), which in humans functions on a cycle of about 24 hours.^8,9 To entrain the biological clock with the environment, it receives a dark/light input via photoreceptors of the eye’s retina known as the non–image forming photoreceptors (NIFPs) containing the photopigment melanopsin.^10,11 Moreover, it results in an output—the neurohormone melatonin (MLT) produced and secreted by the pineal gland, which among others, signals environmental temporal information to organs, tissues, and cells, including the suprachiasmatic nucleus cells. Short wavelength (SWL) illumination at the range of ~440 to 520 nm may efficiently suppress the synthesis of the pineal neurohormone MLT.^12-16 Since 1879 when Thomas Alva Edison invented the commercialized incandescent bulb, commonly used indoor and outdoor lightings contain SWL illumination. However, contemporary fluorescent and blue-white light emitting diode (LED) illuminations from various sources, including outdoor lighting and indoor lighting (such as TV, cellular and smart phones, as well as other electronic and electrical devices) have strong SWL emissions, therefore highly suppressing MLT production.^4,13,17,18 Suppression of MLT production during the night and across multiple years may lead to an increased breast cancer (BC) risk.¹⁹ Various potential mechanisms linking circadian disruption and exposure to ALAN with cancer risks and especially BC were suggested. All authors refer to a higher BC risk resulting from MLT suppression, disruption of sleep/wake cycles that influence the immune system that are MLT independent, alteration in cell cycle, and clock gene function.^7,17,19 Furthermore, the nexus between ALAN and MLT suppression, epigenetic modifications, and BC were also reviewed. It was noted that MLT suppression by ALAN results in an epigenetic modification, global DNA methylation, characterized by losing methyl groups or hypomethylation. However, MLT treatment reversed the process, resulting in recovery of global DNA methylation levels.²⁰

Various cancers are associated with a long interval between exposure to an environmental putative risk factor and consequent diagnosis,²¹ whereas BC probably has a latent period longer than a few years.^22,23 Therefore, the risk for developing BC may rise following women’s exposure to indoor and/or outdoor ALAN years before diagnosis.²⁴

In 2007, the World Health Organization’s International Agency for Research on Cancer published a decision based on experimental and epidemiological studies, which said that “shift work that involves circadian disruption is probably carcinogenic to humans.”^{25 (p. 1065)} This decision refers to occupational exposure but not to environmental exposure. Furthermore, in 2012, the American Medical Association (AMA) acknowledged that “excessive light at night can disrupt sleep, exacerbate sleep disorders and cause unsafe driving conditions.”²⁶

The objective of this study was to compare past indoor and outdoor light exposure habits of BC patients with controls representing the general Israeli-Jewish population, excluding shift workers. As such, we hypothesized that if exposure to ALAN illumination is associated with a higher risk of BC development, then women who had been exposed to higher levels of indoor and/or outdoor light were likely to exhibit, ceteris paribus, higher BC rates, 10 to 15 years after exposure. To test this hypothesis, in order to suggest a model for the nexus between past lighting habits and BC risk, we compared lifestyle and light exposure habits between BC patients and controls from urban and rural areas using a questionnaire.

Materials and Methods

Study Participants and Questionnaire

All participants were recruited between the years 2010 and 2014 and were Israeli-Jewish non–shift workers, 29 to 91 of age. The sample size consisted of cases and controls in a ratio of about 1:2, respectively, to allow expression of significant results regarding differences between the groups and odds ratios (with 95% confidence intervals). Cases were BC patients attending the Comprehensive Cancer Center in Soroka Medical Center (SMC), Beer-Sheva, and the Baruch Padeh, Poria Medical Center (PMC) in Tiberius. BC patients were asked by the medical staff to fill out a questionnaire in the clinic (81%) or at home (19%) according to their choice. In case of questions unclear to them, they were guided to consult the medical staff. In all, 80% of the participants completed the questionnaire. Healthy control participants were friends of BC patients and women recruited through personal meetings in schools. Controls, with no documented BC history or any other type of cancer, were matched with BC patients according to age and residential area. Residential area was determined based on the definition of the Israeli Bureau of Statistics, which states that a settlement with more than 2000 residents is an urban area, whereas a settlement with less than 2000 residents is rural.²⁷ BC patients and controls maintained their residential area during the reference period of the study.

Case and control participants completed a structured questionnaire eliciting information on indoor and outdoor light habits and lifestyle in the past 10 to 15 years prior to diagnosis (cases) or 10 to 15 years prior to the time of the questionnaire (controls). Past light habits and lifestyle can take part in generating disease that has a relatively long latency period, such as BC.^22,23

In the questionnaire, participants were asked to relate their habits referring to a period of 5 years, 10 to 15 years before diagnosis. Although the possibility of imprecise retrieval (recall) of relevant habits exists, there is no other way to learn about those habits in retrospect. Sleep quality, falling asleep with TV on (turning the TV off before sleeping), sleeping with the TV on for most of the night, exposure to outdoor and indoor lighting in the sleeping habitat, use of bed lamps or room lamps for reading before retiring to sleep were variables that were part of the questionnaire. Answers were scaled from 1 to 5, where 1 was lowest and 5 highest. Scale points were the following: 1, never; 2, rarely; 3, sometimes; 4, often; 5, daily (Likert scale). Sleep duration was calculated based on bedtime (“At what time did you go to sleep, and at what time did you wake up?”). Participants were also asked to mark the type of illumination source used as bedroom light and reading light. Pictures of various illumination sources such as fluorescent, halogen, and incandescent bulbs were presented in the questionnaire. Individual data such as age, number of births, BC familial history, hormone therapy and menstruation, eating routine and the extent of a balanced diet, and alcohol and coffee drinking habits, together with geographical data such as country of birth and urban or rural residential area were collected.

Statistical Analysis

Data from questionnaires were analyzed using SPSS19. The χ² test and Student t test were used to determine significant differences between cases and controls with dichotomous variables and continuous variables, respectively. Effect size was reported using Cohen’s d value to indicate the standardized difference between 2 means. The meaning of Cohen’s d effect size is interpreted as follows: we refer to the value of 0.2 as a small effect, about 0.5 a medium effect, and 0.8 to infinity as a large effect.²⁸ Binary logistic regression models were used to obtain a prediction model (ORs, 95% CIs) for BC risk with exposures to ALAN as predictor variables together with confounders. To eliminate bias or confounding, we assembled 2 models. In the first model, we used stepwise logistic regression to examine the relationship between known documented variables such as genetics, reproductive factors, and lifestyle with BC risk. In the second model, we used hierarchal logistic regression to examine variables that emerged out of the first model together with ALAN variables. Because BC has a latency period differing by host or environmental characteristic, we focused on identification of ALAN variables that may explain BC risk. Our purpose was to reveal ALAN hazardous exposures that occurred in a period of 5 years, 10 to 15 years prior to disease detection. In the hierarchal logistic regression, the first block contains the variable “place of birth” (Israel/out of Israel); this is the only variable that emerged out of the first model. The second block contained variables of sleep: sleep duration and sleep quality. The third block contained ALAN variables: subjective light intensity in the bedroom during the night, sleeping with light at an intensity level for reading, reading with room light illumination before retiring to sleep, reading with bed light illumination before retiring to sleep, falling asleep with the TV on (turning the TV off before sleeping), sleeping with the TV on for most of the night, sleeping with light penetrating into the room from outside, sleeping with dim light during the night, sleeping with closed shutters during the night, turning the lights on if waking up during the night, residing near strong ALAN sources, type of bedroom illumination, and type of bed light illumination. The model’s R² (measure of predictive power) is presented using Cox and Snell, with a maximum of less than 1, and Nagelkerke, with a maximum of 1 (for information see Table 3).²⁹ A significance of P < .05 (2 tailed) was set for acceptance.

This study was approved by SMC Helsinki committee and PMC Helsinki committee. Furthermore, it was approved by the ethical board, University of Haifa, Israel. Therefore, all aspects of the study complied with ethical standards.

Results

Descriptive Characteristics

A total of 278 women participated in this study (2010-2014), with 93 (33.5%) BC patients and 185 (66.5%) non-BC controls. Cases and controls were age matched; therefore, there was no significant age difference (t_(232.88) = 0.87) between the groups. The mean age of cases (55.60 ± 9.52 years) was similar to that of controls (54.43 ± 12.49 years). Home residence of cases and controls was also matched. Residential segmentation of BC patients and controls during the time under study was as follows: 22.6% (n = 21) of BC patients lived in rural areas, whereas 77.4% (n = 72) lived in urban areas; 30.8% (n = 57) of controls lived in rural areas, whereas 69.2% (n = 128) lived in urban areas (χ²₍₁₎ = 2.08). For information see Table 1, which displays descriptive statistics of research variables.

Table 1.

Comparison Between Cases and Controls in a 5-Year Period, 10 to 15 Years Before Breast Cancer Diagnosis in Israeli Women.^a

Variable	Case (n = 93)		Control (n = 185)		Statistic Comparing the 2 Groups	Effect Size Cohen’s d
Age (years)	55.60 ± 9.52		54.43 ± 12.49		t_(232.88) = 0.87	—
Residential area	Rural 21, 22.6%	Urban 72, 77.4%	Rural 57, 30.8%	Urban 128, 69.2%	χ²₍₁₎ = 2.08	—
Place of birth	Israel 54, 58.1%	Other 39, 41.9%	Israel 133, 71.9%	Other 52, 28.1%	χ²₍₁₎ = 5.37*, φ = 0.14	—
Sleep duration (hours)	7:47 ± 1:05		8:04 ± 1:05		t₍₂₇₆₎ = 2.14*	0.27
The number of times awoken at night	2.11 ± 1.21		2.00 ± 1.12		t₍₂₇₆₎ = 0.73	—
Subjective sleep quality	3.63 ± 1.39		3.76 ± 1.25		t₍₂₇₆₎ = 0.83	—
Subjective light intensity in the bedroom during the night	2.29 ± 0.70		2.17 ± 0.74		t₍₂₇₆₎ = 1.33	—
Sleeping with light in an intensity for reading	1.57 ± 1.13		1.59 ± 1.14		t₍₂₇₆₎ = 0.11	—
Reading with room light illumination before retiring to sleep	2.41 ± 1.52		2.45 ± 1.35		t_(166.77) = 0.22	—
Reading with bed light illumination before retiring for sleep	2.39 ± 1.62		2.88 ± 1.56		t₍₂₇₆₎ = 2.44*	0.31
Falling asleep with the TV on	2.18 ± 1.53		2.22 ± 1.42		t₍₂₇₆₎ = 0.21	—
Sleeping with the TV on for the most of the night	1.74 ± 1.29		1.66 ± 1.12		t₍₂₇₆₎ = 0.54	—
Sleeping with light penetrating the room from outside	2.76 ± 1.78		2.69 ± 1.53		t_(161.97) = 0.35	—
Sleeping with dim light in the room during the night	1.55 ± 1.21		1.56 ± 1.12		t₍₂₇₆₎ = 0.13	—
Sleeping with closed shutters during the night	2.45 ± 1.58		2.89 ± 1.47		t₍₂₇₆₎ = 2.28*	0.29
Turning the lights on when waking up during the night	2.25 ± 1.55		2.53 ± 1.40		t₍₂₇₆₎ = 1.51	—
Resides near strong ALAN sources	1.53 ± 1.26		1.18 ± 0.61		t_(114.02) = 2.56**	0.35
Type of bedroom illumination LWL/SWL	LWL 72, 77.4%	SWL 21, 22.6%	LWL 159, 85.9%	SWL 26, 14.1%	χ²₍₁₎ = 3.20^, φ = 0.11	—
Type of bed light illumination LWL/SWL	LWL 81, 87.1%	SWL 12, 12.9%	LWL 172, 93.0%	SWL 13, 7.0%	χ²₍₁₎ = 2.61	—

Abbreviations: ALAN, artificial light at night; LWL, long wavelength; SWL, short wavelength.

^P < .1, *P < .05, **P < .01. Variables in bold are significantly different; analysis is based on a sample of n = 93 cases and 185 controls. Data refer to habits from a 5-year period, 10 to 15 years before diagnosis.

Few significant differences between cases and controls were revealed in this study. A significant (χ²₍₁₎ = 5.37;* p < 0.05; φ = 0.14) difference and a positive association was noted between the groups regarding place of birth: 58.1% (n = 54) of the BC patients were born in Israel, whereas 41.9% (n = 39) were immigrants; 71.9% (n = 133) of the controls were born in Israel, whereas only 28.1% (n = 52) were immigrants.

Significant (t₍₂₇₆₎ = 2.14,* p<0.05 Cohen’s d = 0.27) differences between the groups were also revealed in sleep duration. BC patients used to sleep less (7:47 ± 1:05) than controls (8:04 ± 1:05) during the studied period. Moreover, exposure to ALAN inside the sleeping habitat while reading with bed light illumination before retiring to sleep differed significantly (t₍₂₇₆₎ = 2.44; *p < 0.05; Cohen’s d =0.31) between cases and controls. Cases used to read rarely using bed light illumination (2.39 ± 1.62), whereas controls sometimes used bed light illumination before retiring to sleep (2.88 ± 1.56). Reducing light penetrating from the outside using closed shutters while sleeping was also significantly different between BC patients and controls (t₍₂₇₆₎ = 2.28; *p < 0.05; Cohen’s d = 0.29). Patients used to sleep less often (rarely) with closed shutters (2.45 ± 1.58), whereas controls used closed shutters sometimes (2.89 ± 1.47). The mean exposure of BC patients as a result of living near strong ALAN sources in urban and rural areas was 1.53 ± 1.26, whereas exposure of controls as a result of living near strong ALAN sources in the same residential areas was significantly smaller, with a mean value of 1.18 ± 0.61 (t_(114.02) = 2.56; **p < 0.01; Cohen’s d = 0.35).

Other variables referring to indoor and outdoor exposures of BC patients and controls did not differ significantly in this study (data presented in Table 1). The variables associated with number of births, BC familial history, hormone therapy and menstruation, eating routine and the extent of a balanced diet, and alcohol and coffee drinking were examined, and no significant differences were revealed between the groups (data not shown).

Model Analysis

Analysis included the assembly of 2 models one after the other. In the first model, we used stepwise logistic regression to examine the relationship between known documented variables such as genetics, reproductive factors, and lifestyle with BC risk. OR of the tested variables was not significant (data not shown) except for place of birth (OR = 0.54; 95% CI = 0.32-0.91; P < .02). In the second model, we used place of birth together with sleep and ALAN variables. OR results of binary logistic regression reveal that women who slept longer in a 5-year period, 10-15 years before diagnosis, reduced their BC risk significantly (OR = 0.74; 95% CI = 0.57-0.97; P < .03). Likewise, women who were exposed to ALAN as a result of reading with bed light (reading lamp) illumination reduced their BC risk (OR = 0.81; 95% CI = 0.67-0.97; P < .02); also, sleeping with closed shutters lowered the risk for BC development (OR = 0.82; 95% CI = 0.68-0.99; P < .04). However, women who were exposed to ALAN as a result of living near strong ALAN sources were at a significantly higher (OR = 1.52; 95% CI = 1.10-2.12; P < .01) BC risk.

The model summary in 3 steps is presented in Table 3. Accumulative explained variance of all the variables in the model (step 3) is 17% (χ²₍₁₇₎ = 36.57; **p < 0.01; Nagelkerke R² = 0.17; Cox and Snell R² = 0.12). Personal characteristics (step 1), which includes place of birth, explains 3% of the variance (χ²₍₁₎ = 5.28; *p < 0.05; Nagelkerke R² = 0.03; Cox and Snell R² = 0.02). Sleep habits (step 2) explains 2% of the variance. Total variance for step 2 is 5% (χ²₍₄₎ = 10.91; *p < 0.05; Nagelkerke R² = 0.05; Cox and Snell R² = 0.04). Adding variables representing exposure to ALAN (step 3) contributes 12% more to the variance, with a total of 17% (data presented above). Significant OR results were obtained for the following variables: sleep duration, reading with bed light illumination before retiring to sleep, sleeping with closed shutters during the night, and residing near strong ALAN sources (data presented in Table 2).

Table 2.

OR of Variables Affecting Breast Cancer Risk in Israeli Women (Binary Logistic Regression) 10 to 15 Years Before Diagnosis.^a

Variable	P Value	OR (95% CI)
Sleep duration	.03	0.74 (0.57-0.97)
Reading with bed light illumination before retiring to sleep	.02	0.81 (0.67-0.97)
Sleeping with closed shutters during the night	.04	0.82 (0.68-0.99)
Resides near strong ALAN sources	.01	1.52 (1.10-2.12)

Abbreviations: OR, odds ratio; ALAN, artificial light at night.

Based on a sample of n = 93 cases and 185 controls.

Table 3.

Hierarchical Binary Logistic Regression Analysis of Variables Affecting Breast Cancer Risk in Israeli Women, 10 to 15 Years Before Diagnosis.^a

Step	1		2		3
χ²	5.28*		10.91*		36.57**
Nagelkerke R²	0.03		0.05		0.17
Cox and Snell R²	0.02		0.04		0.12
Predictors	Exp (B)	Wald	Exp (B)	Wald	Exp (B)	Wald
Place of birth Israel/out of Israel	0.54	5.31*	0.54	5.14*	0.65	2.21
Sleep duration			0.76	4.79*	0.74	4.64*
The number of times awoken at night			0.99	0.003	1.09	0.32
Subjective sleep quality			0.89	0.94	1.00	0.001
Subjective light intensity in the bedroom during the night					1.21	0.78
Sleeping with light at the intensity for reading					0.96	0.1
Reading with room light illumination before retiring to sleep					0.96	0.17
Reading with bed light illumination before retiring to sleep					0.81	5.23*
Falling asleep with the TV on					0.84	1.80
Sleeping with the TV on for most of the night					1.26	1.91
Sleeping with light penetrating the room from outside					0.96	0.16
Sleeping with dim light during the night					0.89	0.82
Sleeping with closed shutters during the night					0.82	4.23*
Turning the lights on when waking up during the night					0.88	1.69
Residing near strong ALAN sources					1.52	6.33**
Type of bedroom illumination: LWL/SWL					1.35	0.5
Type of bed light illumination: LWL/SWL					1.56	0.64

Abbreviations: OR, odds ratio; ALAN, artificial light at night; LWL, long wavelength; SWL, short wavelength.

*P < .05; **P < .01. Variables in bold have a significant OR. Based on a sample of n = 93 cases and 185 controls. Data refer to habits from a 5-year period, 10 to 15 years before diagnosis.

Discussion

BC is increasing worldwide in developed and developing countries.³⁰ The results of several studies^{7,17,19,24,31} try to explain this rise with the increase of ALAN levels. The analysis of 278 Israeli women in our current study, regarding habits in a 5-year period 10 to 15 years before diagnosis, reveals a significant correlation between ALAN and BC. The results of our study revealed that BC risk increased with an increased exposure to light as a result of residence near strong ALAN sources. Nevertheless, BC risk decreased with longer night sleep, decreased exposure to light penetrating from the outside as a result of closed shutters, and reading with bed light illumination before retiring to sleep. These associations were still present after controlling for age, urban/rural residential area, and other potential confounders, which were measured. Our results, suggesting that higher BC risk is associated with residence near strong ALAN sources, are consistent with those of others^31-33 claiming that women living in areas with high ALAN levels may have increased BC risk. However, our results are based on a self-reported, individual-level questionnaire rather than on imagery data such as that of the US Defense Meteorological Satellite Program (DMSP). Moreover, our results refer to local exposures in a period of 5 years, 10 to 15 years before diagnosis (which may be the estimated latency period for BC development²¹) and, therefore, differ in terms of varying methods and time spans used with DMSP data. The fact that our findings persisted after controlling for residential area provides reasonable assurance that the observed increased OR does not reflect the higher BC rates in urban areas alone.^34,35 An individual’s exposure to outdoor lighting depends on natural and artificial illumination of the environment. Therefore, it is challenging to influence this exposure. On the other hand, indoor light exposures are more likely to be influenced by an individual’s lifestyle and habits. Still, both indoor and outdoor light exposures may negatively influence MLT production levels and, therefore, increase BC risk.^17,36-39 Our results regarding indoor light exposures and sleeping habits reveal that BC risk decreased with longer night sleep duration for controls 10 to 15 years before the time of the study, even after controlling for age. The difference in sleep duration between cases and controls, although significant, is less than 20 minutes and within the considered optimum sleep duration for most adults.⁴⁰ Nevertheless, this documented time gap may correspond to the latency period, and therefore, it may cautiously indicate the beginning of degradation in health.²² Results from the literature on BC and sleep vary: whereas most studies report no correlation with sleep duration,⁴¹ there is limited evidence that suggests that longer sleep durations may be related to increase in BC risk, depending on BC subtype⁴² and the format of assessing sleep duration.^43-45 Our results are consistent with those obtained in other studies (as in Verkasalo et al⁴⁶), suggesting a decreased BC risk for long sleepers. Although there are differences between the structures of both studies, the results are similar. Results that suggest an association between BC and sleep duration may emerge from the following: (1) exposure to light beyond a threshold in terms of light intensity, wavelength, and duration that occurred in the past may have caused a change in sleep duration, thus, affecting the health condition; (2) a change in health (not documented) led to a change in sleep duration; (3) a change in sleep duration led to a change in exposure to ALAN and health conditions.

An additional ALAN variable that was explored in this study is sleeping with closed shutters in the bedroom. Our results show a significant decrease in BC risk with more frequent use of shutters by the controls. Decreasing light penetrating from the outside into the sleeping habitat using closed shutters was discussed by Kloog et al³³ with the same trend. The results of a previous study by Davis et al²⁴ suggested an indication of increased BC risk for patients with the brightest bedrooms, without specifying the source of light. All the same, most of the modern lighting used outdoors, such as metal halide, mercury vapor, fluorescent lamps, and LEDs, contain SWL light, which, when penetrating the sleeping habitat, even at low intensities, can inhibit MLT production and secretion, thus influencing human physiology and increasing BC risk.^11,12,18,33 A decreased BC risk was also noted, to the best of our knowledge for the first time in our study, among participants who reported reading with bed light illumination before retiring to sleep. Using bed light illumination is part of indoor lighting habits and exposures. Bed lamps are typically low-intensity, indirect shielded lamps. Using bed lamps is possibly a part of a routine before going to sleep and one of its possible advantages is that the light source is not directed to the retina but rather to the printed material. The results of a former study conducted by Gooley et al³⁷ revealed that exposure to room light transmitted through a UV stable filter (<200 lux) before bedtime, compared with dim light (<3 lux), suppresses MLT onset and shortens MLT production and secretion duration by about 90 minutes in healthy participants. The authors also state that exposure to room light through the regular hours of sleep suppressed plasma MLT levels by more than 50% in most trials. Therefore, habitual exposure to ALAN may lead to a shortened duration of MLT production, delayed sleep, and delayed onset of advanced stages such as rapid eye movement.^37,47 These changes in MLT levels can potentially influence body physiology and health including BC.^25,37,48 These results obtained by exposure to dim light may support the results of our current study showing a correlation between the use of bed light illumination and a decreased risk of BC. Based on our results, we suggest that using bed light illumination, presumably long wavelength (LWL), with a dominant wavelength greater than ~550 nm, does not interfere with MLT production and secretion, as shown by Cajochen et al¹³ and, therefore, does not have a negative effect. Most of our participants used LWL incandescent/halogen illumination as bed light (BC patients 87.1% and controls 93%) or room light (BC patients 77.4% and controls 85.9%). Therefore, the exposure was minimized as well as the negative effect on MLT production, similar to red dim light.^4,49 However, we consider that those results include habits from an era before smartphones and iPads; SWL portable illumination that can easily penetrate the sleeping habitat was less common 10 to 15 years ago.

Limitations of this study include the following: (1) Possible difficulties of the participants in retrieving answers to questions referring to habits 10 to 15 years (a period of 5 years) before the time of the study could have caused bias and, therefore, may have influenced the relationship between nighttime light habits and BC incidence. (2) Nighttime light habits were determined by self-report; however, participants were encouraged to record their habits referring to some reference points in their past, such as age, age of children, working and dwelling places, and so on, in order to strengthen their selections of codes. (3) Regarding type of illumination, we presented to the participants relevant figures and encouraged them to consult other family members in order to supply a correct answer. (4) Sample composition was based on BC patients in 2 different hospitals in Israel; it was difficult to recruit enough BC participants in Beer Sheva with matching controls. Therefore, we recruited 32 more in Poria medical center. However, urban or rural alignment was similar and followed the definition of the Israeli Bureau of Statistics. Moreover, the recruitment of a larger control group compensated for the recruitment of cases from 2 medical centers. (5) There is always the possibility of residual confounding that cannot be excluded. In this study, light variables were prominent over other potential confounders. Despite these limitations, the fact that the association between exposure to ALAN during the scotophase in a 5-year period, 10 to 15 years earlier, and an increased risk for BC remains even after controlling for other silent variables suggest that our nighttime light habits in the past may be a significant BC risk factor.

In conclusion, our findings show that measured use of indoor and outdoor ALAN may lower BC risk relative to excessive use of ALAN. Pineal MLT production is sensitive to SWL lighting, and as a result, past personal lighting habits can be a part of BC etiology. With growing evidence that ALAN exposure as well as the growing use of LED electronic devices with SWL light can be harmful to health, we suggest that future studies determining the impact of chronic nighttime exposure to ALAN on MLT suppression and BC risk in humans are needed. Moreover, the results of this study can be part of an evidence base that should lead decision makers to consider preventive actions, including education of the younger generation regarding reduction in indoor and/or outdoor ALAN pollution and exposure.

Footnotes

Acknowledgements

We thank Dr Michael Koretz of the Elisheva Kaplan Eshkol Comprehensive Breast Health Center of Soroka University Medical Center, Beer-Sheva, for allowing us to approach BC patients and controls visiting the clinic and Dr David Gefen and Dr Rita Toker of the Oncology Day Care Clinic of the Soroka University Medical Center, Beer-Sheva, for helping in handing out the questionnaires to BC patients.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Gaston

Duffy

Gaston

Bennie

Davies

TW.

Human alteration of natural light cycles: causes and ecological consequences. Oecologia. 2014;176:917-931.

Arendt

Melatonin and the pineal gland: influence on mammalian seasonal and circadian physiology. Rev Reprod. 1998;3:13-22.

Holker

Wolter

Perkin

Tockner

Light pollution as a biodiversity threat. Trends Ecol Evol. 2010;25:681-682.

Falchi

Cinzano

Elvidge

Keith

Haim

Limiting the impact of light pollution on human health, environment and stellar visibility. J Environ Manage. 2011;92:2714-2722.

Gaston

Visser

Holker

. The biological impacts of artificial light at night: the research challenge [published online May 5, 2015]. Philos Trans R Soc Lond B Biol Sci. doi:10.1098/rstb.2014.0133.

Pevet

Challet

Melatonin: both master clock output and internal time-giver in the circadian clocks network. J Physiol Paris. 2011;105:170-182.

Blask

DE.

Melatonin, sleep disturbance and cancer risk. Sleep Med Rev. 2009;13:257-264.

Czeisler

Duffy

Shanahan

. Stability, precision, and near-24-hour period of the human circadian pacemaker. Science. 1999;284:2177-2181.

Ralph

Foster

Davis

Menaker

Transplanted suprachiasmatic nucleus determines circadian period. Science. 1990;247:975-978.

10.

Thapan

Arendt

Skene

DJ.

An action spectrum for melatonin suppression: evidence for a novel non-rod, non-cone photoreceptor system in humans. J Physiol. 2001;535:261-267.

11.

Brainard

Hanifin

Greeson

. Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor. J Neurosci. 2001;21:6405-6412.

12.

Pauley

SM.

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

13.

Cajochen

Munch

Kobialka

. High sensitivity of human melatonin, alertness, thermoregulation, and heart rate to short wavelength light. J Clin Endocrinol Metab. 2005;90:1311-1316.

14.

Lewy

Wehr

Goodwin

Newsome

Markey

Light suppresses melatonin secretion in humans. Science. 1980;210:1267.

15.

Morita

Tokura

Effects of lights of different color temperature on the nocturnal changes in core temperature and melatonin in humans. Appl Human Sci. 1996;15:243-246.

16.

Morita

Teramoto

Tokura

Inhibitory effect of light of different wavelengths on the fall of core temperature during the nighttime. Jpn J Physiol. 1995;45:667-671.

17.

Haim

Portnov

BA.

Light Pollution as a New Risk Factor for Human Breast and Prostate Cancers. New York, NY: Springer; 2013.

18.

Kerenyi

Pandula

Feuer

. Why the incidence of cancer is increasing: the role of “light pollution.” Med Hypotheses. 1990;33:75-78.

19.

Haus

Smolensky

MH.

Shift work and cancer risk: potential mechanistic roles of circadian disruption, light at night, and sleep deprivation. Sleep Med Rev. 2013;17:273-284.

20.

Haim

Zubidat

. Artificial light at night: melatonin as a mediator between the environment and epigenome [published online May 5, 2015]. Philos Trans R Soc Lond B Biol Sci. doi:10.1098/rstb.2014.0121.

21.

McPherson

Coope

Vessey

MP.

Early oral contraceptive use and breast cancer: theoretical effects of latency. J Epidemiol Community Health. 1986;40:289-294.

22.

MacMahon

Cole

Brown

Etiology of human breast cancer: a review. J Natl Cancer Inst. 1973;50:21-42.

23.

Armenian

HK.

Incubation periods of cancer: old and new. J Chronic Dis. 1987;40:9S-15S.

24.

Davis

Mirick

Stevens

RG.

Night shift work, light at night, and risk of breast cancer. J Natl Cancer Inst. 2001;93:1557-1562.

25.

Straif

Baan

Grosse

. Carcinogenicity of shift-work, painting, and fire-fighting. Lancet Oncol. 2007;8:1065-1066.

26.

American Medical Association. AMA adopts new policies at annual meeting. https://download.ama-assn.org/resources/doc/csaph/x-pub/a12-csaph4-lightpollution-summary.pdf

27.

Israel central bureau of statistics. http://www.cbs.gov.il/ishuvim/ishuv2013/info2013.pdf

28.

Cohen

Statistical Power Analysis for the Behavioral Sciences (2nd ed). Hillsdale, NJ: Erlbaum; 1988.

29.

Menard

Coefficients of determination for multiple logistic regression analysis. Am Stat. 2000;54:17-24.

30.

Jemal

Bray

Center

Ferlay

Ward

Forman

Global cancer statistics. CA Cancer J Clin. 2011;61:69-90.

31.

Hurley

Goldberg

Nelson

. Light at night and breast cancer risk among California teachers. Epidemiology. 2014;25:697-706.

32.

Bauer

Wagner

Burch

Bayakly

Vena

JE.

A case-referent study: light at night and breast cancer risk in Georgia. Int J Health Geogr. 2013;12:23.

33.

Kloog

Portnov

Rennert

Haim

Does the modern urbanized sleeping habitat pose a breast cancer risk?. Chronobiol Int. 2011;28:76-80.

34.

Doll

Urban and rural factors in the aetiology of cancer. Int J Cancer. 1991;47:803-810.

35.

Reynolds

Hurley

Goldberg

. Regional variations in breast cancer among California teachers. Epidemiology. 2004;15:746-754.

36.

Czeisler

Allan

Strogatz

. Bright light resets the human circadian pacemaker independent of the timing of the sleep-wake cycle. Science. 1986;233:667-671.

37.

Gooley

Chamberlain

Smith

. Exposure to room light before bedtime suppresses melatonin onset and shortens melatonin duration in humans. J Clin Endocrinol Metab. 2010;96:E463-E472.

38.

Stevens

Hilakivi-Clarke

Alcohol exposure in utero and breast cancer risk later in life. Alcohol Alcohol. 2001;36:276-277.

39.

Vollmer

Michel

Randler

Outdoor light at night (LAN) is correlated with eveningness in adolescents. Chronobiol Int. 2012;29:502-508.

40.

Centers for Disease Control and Prevention. Insufficient sleep is a public health epidemic. http://www.cdc.gov/features/dssleep/. Accessed November 12, 2015.

41.

Qin

Zhou

Zhang

Wei

Sleep duration and breast cancer risk: a meta-analysis of observational studies. Int J Cancer. 2014;134:1166-1173.

42.

Vogtmann

Levitan

Hale

. Associa-tion between sleep and breast cancer incidence among postmenopausal women in the Women’s Health Initiative. Sleep. 2013;36:1437-1444.

43.

Hurley

Goldberg

Bernstein

Reynolds

Sleep duration and cancer risk in women. Cancer Causes Control. 2015;26:1037-1045.

44.

McElroy

Newcomb

Titus-Ernstoff

Trentham-Dietz

Hampton

Egan

KM.

Duration of sleep and breast cancer risk in a large population-based case-control study. J Sleep Res. 2006;15:241-249.

45.

Pinheiro

Schernhammer

Tworoger

Michels

KB.

A prospective study on habitual duration of sleep and incidence of breast cancer in a large cohort of women. Cancer Res. 2006;66:5521-5525.

46.

Verkasalo

Lillberg

Stevens

. Sleep duration and breast cancer: a prospective cohort study. Cancer Res. 2005;65:9595-9600.

47.

Takeuchi

Uchimura

Hashizume

. Melatonin therapy for REM sleep behavior disorder. Psychiatry Clin Neurosci. 2001;55:267-269.

48.

Blask

Brainard

Dauchy

. Melatonin-depleted blood from premenopausal women exposed to light at night stimulates growth of human breast cancer xenografts in nude rats. Cancer Res. 2005;65:11174-11184.

49.

Burgess

Molina

TA.

Home lighting before usual bedtime impacts circadian timing: a field study. Photochem Photobiol. 2014;90:723-726.