Does the modern urbanized sleeping habitat pose a breast cancer risk?

Chronobiology International, 28(1): 76–80, (2011)Copyright Informa Healthcare USA, Inc.
ISSN 0742-0528 print/1525-6073 onlineDOI: 10.3109/07420528.2010.531490 Does the Modern Urbanized Sleeping Habitat Pose a Breast Cancer Risk? Itai Kloog,1,4 Boris A. Portnov,1 Hedy S. Rennert,2 and Abraham Haim3 1Department of Natural Resources and Environmental Management, Graduate School of Management, University of Haifa,Haifa, Israel, 2Department of Community Medicine and Epidemiology, Carmel Medical Center and Bruce Rappaport Faculty ofMedicine, Technion–Israel Institute of Technology and Clalit Health Services National Cancer Control Center, Haifa, Israel, 3TheIsraeli Center for Interdisciplinary Research in Chronobiology, University of Haifa, Haifa, Israel, 4Department of EnvironmentalHealth—Exposure, Epidemiology and Risk Program, Harvard School of Public Health, Boston, Massachusetts, USA Due to its disruptive effects on circadian rhythms and sleep deprivation at night, shiftworking is currently recognized asa risk factor for breast cancer (BC). As revealed by the present analysis based on a comparative case-control study of1679 women, exposure to light-at-night (LAN) in the “sleeping habitat” is significantly associated with BC risk (oddsratio [OR] = 1.220, 95% confidence interval [CI] = 1.118–1.311; p < .001), controlling for education, ethnicity, fertility,and alcohol consumption. The novelty of the present research is that, to the best of the authors’ knowledge, it isthe first study to have identified an unequivocal positive association between bedroom-light intensity and BC risk.
Thus, according to the results of the present study, not only should artificial light exposure in the workingenvironment be considered as a potential risk factor for BC, but also LAN in the “sleeping habitat.” (Authorcorrespondence: [email protected]) Keywords: Breast cancer, Circadian disruption, Light-at-night, Light pollution, Sleeping habitat and affordable for wide segments of the population, Several possible mechanisms have been suggested in resulting in increased LAN exposure in both outdoor, recent years to explain the association between exposure such as public areas and transportation, and indoor to light pollution, termed “light-at-night” (LAN), and spaces, such as workplaces, apartments, and houses.
breast cancer (BC). These mechanisms include the sup- Although several empirical studies have been conducted pression of melatonin (MLT) secretion by the pineal to evaluate the effects of work-related and outdoor light gland leading to increased tumor growth (Blask et al., pollution on hormone-dependent cancers (Davis et al., 2005) and the adverse effects of LAN on thermoregula- 2001; Kloog et al., 2008, 2009a, 2009b, 2010; Schernham- tory (Haim et al., 2005) and immune functions (Nelson, mer et al., 2006), the association between LAN exposure 2004; Stevens et al., 2007). The “LAN-BC” theory is sup- in the bedroom of apartments and houses (which we ported by the results of several recent epidemiological term the “sleeping habitat”) and cancer has largely Chronobiol Int Downloaded from informahealthcare.com by 85.250.145.131 on 12/23/10 studies carried out at various population levels (Kloog escaped investigation, and the existing evidence is essen- et al., 2008, 2009a, 2010) that indicated exposure to tially scarce. In the framework of their Long Island study, LAN was associated with elevated BC rates, whereas no O’Leary and colleagues (2006) conducted personal inter- such an association was found for lung cancer. Other views to trace LAN-exposure histories at work and home.
studies provided further evidence that women working The study cohort included 576 women diagnosed with night shifts are at high risk of developing BC (Davis BC and 585 population-based controls. The question- et al., 2001; Hansen, 2001; Schernhammer et al., 2001), naire included questions about the number of sleep whereas studies on blind women have found BC rates hours, frequency of turning lights-on during the night, to be lower than those of visually unimpaired women and the length of time the light was on during the night- in the same populations (Feychting et al., 1998; Kliukiene time, but it did not include questions about bedroom- et al., 2001; Verkasalo et al., 1999).
light intensity. The analysis of the data revealed that With accelerating urbanization worldwide and rising women who frequently turned on lights during the living standards, electricity has become more accessible sleep hours exhibited an increased BC risk (odds ratio Submitted June 1, 2010, Returned for revision June 2, 2010, Accepted October 2, 2010 Address correspondence to: Prof. Abraham Haim, The Israeli Center for Interdisciplinary Research in Chronobiology, University of Haifa,Mount Carmel, Haifa, Israel 31905. Tel.: 972-4-8288784; E-mail: [email protected] The Sleeping Habitat and Breast Cancer Risk [OR] = 1.65, 95% confidence interval [CI]: 1.02–2.69). In et al., 2008). Instead of predetermining the number of another study, Davis and colleagues (2001) investigated cases and controls to be selected within each population whether exposure to LAN at home is associated with an stratum (stratified sampling), frequency matching in- increased risk of BC in women. The study included volves the selection of cases at random, with controls 813 cancer patients aged 20–74 yrs and 793 control being taken from the corresponding subgroups in subjects of comparable age. Personal interviews were proportion to the number of cases (Rothman et al., conducted to gather information on sleep habits and 2008). Controls were randomly selected from the list of the bedroom-light environment 10 yrs before diagnosis women enrolled in the health-care program provided and lifetime occupational history. Conditional logistic by the Clalit Health Services, the largest health-care pro- regression was then used to adjust for other potential vider in Israel, which covers approximately 60% of the risk factors, such as family history of cancer, parity, country’s adult population. Health-care coverage in Israel is mandatory, and all study participants (both therapy. Although the analysis revealed that BC risk was patients and controls) had similar health-insurance significantly higher among subjects with sleep disturb- ance (OR = 1.14, 95% CI = 1.01–1.28), no clear association Cases and controls were matched by age, location of between bedroom-light intensity and BC was found (OR primary clinic, and ethnicity—Jewish versus non-Jewish (Table 1). Respondents reporting any previous BC were In the past decade, electricity consumption and LAN excluded. All participants were individually interviewed to obtain data, including reproductive and general sources have been introduced, which include the medical history, alcohol consumption, LAN exposure, modern outdoor high-intensity discharge (HID) and flu- and socioeconomic status in order to include these con- orescent lamps that emit blue wavelengths (Pauley, founders in our model. The overall study response rate 2004). These new developments may have presumably was 85.6% of the approached cases and 51.6% among strengthened the LAN-BC association. In the present controls. All participants signed a consent form approved study, we compared light intensities in the “sleeping by the Carmel Medical Center’s institutional review habitat” of 794 BC patients with light intensities at board committee, and all aspects of the study complied homes of a control group of 885 women residing in north- with the ethical standards of the journal (Portaluppi ern Israel. Our working hypothesis was as follows: If exposure to LAN elevates BC risk, then BC patientsshould have been exposed to higher LAN levels in their“ sleeping habitat” than those without BC of the control LAN exposure was evaluated from individual interviews group, controlling for individual level confounders.
regarding exposure to both LAN from outside sourcesand from sources within the households (bedroom- light levels, light coming from outside the bedroom, availability of shutters in the bedroom, and sleepingwith the television on. Nighttime bedroom-light level Data on 1679 women (794 women with BC and 885 con- was evaluated using a 4-point scale: from completely trols) were obtained from the “Breast Cancer in Northern dark (score of 1) to strong light (score of 4). The exact Israel” study, initiated in 2000 and which focused on the wording of the question was as follows: “How do you molecular and environmental etiology of BC in Northern define your nighttime bedroom-light level? “1” (comple- Chronobiol Int Downloaded from informahealthcare.com by 85.250.145.131 on 12/23/10 Israel (Rennert et al., 2010). Northern Israel is formed by tely dark), “2” (low light), “3” (average light), or “4” (very two administrative districts—the Haifa district and the strong light—all lights switched on). Other LAN Northern district—with a combined population of over exposure-related questions included the availability of 2,000,000 residents (ICBS, 2010). In the Haifa district, bedroom shutters and sleeping with the television left about 71% of the total population are Jewish, while 29% on. The answers to these questions were coded dichoto- are Arabs and other ethnic minorities; whereas in the mously, that is, yes or no. Other questions in the ques- Northern district, 44.2% of the population are Jewish, tionnaire referred to alcohol consumption, number of while the rest (55.8%) are Arab Muslims and other births, religion, age, and education.
ethnic minorities (ICBS, 2010). All women residing inthe region and diagnosed with BC since 2000 wereinvited to participate in the study. Questions on LAN exposure were added to the general study questionnaire Statistical analysis was performed using SPSS17TM in mid-2006, and the present study covers 1679 women software and the probability p < .05 (two-sided) was set (794 cases and 885 controls) who responded to the “ex- as the accepted level of statistical significance. Contin- tended” questionnaire between 2006 and 2008.
gency tables, t tests, and unconditional binary logistic Frequency matching, which was used in the present regressions were then used to assess the association study, is a variation of stratified sampling that is between exposure to LAN and BC risk, with adjustment commonly used in empirical studies (cf., e.g., Rothman TABLE 1. Descriptive statistics of the research variables *Mean value and standard deviation in the parentheses, number of cases (percent of total) for all other cases.
†Statistically significant differences between groups by t test.
significant (OR = 0.737, 95% CI = 0.556–0.977, p < .05, and OR = 0.932, 95% CI = 0.886–0.979, p < .01, respect- Table 1 presents the descriptive statistics of the research ively), implying that Jewish women (who exhibit much variables. Table 2 reports the results of the binary logistic lower birth rates—2.72 births for Jews versus 4.07 for regressions run for the whole sample, and Table 3 reports Arabs) are at higher risk of developing BC than non- the results of the binary logistic regressions run for the Jewish survey participants. Other variables, e.g., alcohol Jewish women separately to allow for more homogeneity consumption, education, etc., did not emerge as being significant determinants (p > .05).
As Table 2 shows, the intensity of bedroom light The results of the regression analysis run for Jewish emerged as the strongest predictor of BC, which appears women separately (Table 3) were essentially similar to Chronobiol Int Downloaded from informahealthcare.com by 85.250.145.131 on 12/23/10 to increase its risk significantly (OR = 1.220, 95% the results obtained for the entire sample (Table 2), CI = 1.118–1.311; p < .001). Religion (Jews versus non- with the exception of one variable—alcohol consump- Jews) and number of births also emerged as statistically tion, which emerged as statistically significant (p < .05).
TABLE 2. Factors affecting BC incidence (method: binary logisticregression)* TABLE 3. Factors affecting BC incidence (method: binary logisticregression; Jewish population only)* *Total number of observations = 1679 (794 cases/885 controls).
*Total number of observations = 1277 (609 cases/668 controls).
The Sleeping Habitat and Breast Cancer Risk The protective effect of this factor (OR = 0.688, 95% photoperiod regimes (Haim et al., 1983), light intensities, CI = 0.507–0.935) may be attributed to the fact that and wavelengths (Zubidat et al., 2009, 2010). Therefore, alcohol consumption in Israel, in general, is low, our vision system, which evolved from a mammalian especially among women, and is mostly limited to red system that entailed a subterranean stage in evolution wine, as common for most non-Muslim countries of (Vaughan, 1972), became more complex as it includes the Mediterranean (Kesteloot, 2004). Several studies more types of functional photoreceptors whereby the have shown that red-wine consumption in small quan- NIFPs seem to play a major role in regulating circadian tities can exert a protective effect against the develop- ment of BC (Damianaki et al., 2000; Soleas et al., 2002).
It is interesting to note that recent reviews on MLT, LAN, and BC (Haus, 2009; Srinivasan et al., 2008)report that nighttime MLT treatment in physiological concentrations acts like an endogenous antiestrogen The present analysis is based on a comparative case- that can decrease the formation of estrogens from andro- control study of 1679 women (794 cases and 885 controls) gens through inhibition of the aromatase enzyme. This who were individually interviewed about their lifestyle finding suggests the importance of MLT in modulating and LAN exposure. The study revealed that exposure to estrogen levels in relation to BC risk.
LAN in the “sleeping habitat” was significantly associated The International Agency for Research on Cancer with BC, controlling for education, ethnicity, fertility, and (IARC) recently recognized shiftworking as a 2A probable alcohol consumption. Our results are thus consistent human carcinogen risk factor (Straif et al., 2007). Shift- with those of previous studies that revealed a significant workers are exposed to high LAN intensity in their work LAN-BC association (Kloog et al., 2008, 2009a, 2009b, environment during the dark period of the 24-h cycle, 2010; Schernhammer et al., 2001). Our results also corre- when pineal MLT should be produced and secreted spond with those of the case-control studies conducted into the plasma, thus causing circadian rhythm disrup- by O’Leary and colleagues (2006) and Davis and col- tion. This suppression of pineal MLT production may leagues (2001). However, in the research reported here, lead to increased tumor growth of hormone-dependent the association between bedroom-light intensity and BC ( p < .001) was stronger than it was in the study by The novelty of the present research lies in the fact that Davis and colleagues (2001), thus supporting our to the best of our knowledge, this is the first large-scale working hypothesis of a linkage between LAN exposure case-control study of the general population that has identified a significant positive association between This difference in results may have several expla- bedroom light (“sleeping habitat”) and ambient night- nations. First, our study population was relatively hom- time light (light pollution) levels with BC risk, providing ogenous, with a separate analysis performed for Jewish evidence that the relative risk of BC appears to increase women, whereas the study by Davis and colleagues in more illuminated sleeping environments. Our main (2001) involved a more heterogeneous general popu- conclusion is that not only should LAN in the working lation of diverse ethnic background. Second, it should environment be considered a potential BC risk factor, be borne in mind that the study by Davis and colleagues as noted in shiftworker studies (Hansen, 2001), but also (2001) was conducted some 15 yrs ago (in 1992–1995).
the modern human “sleeping habitat” with high LAN Since then, light pollution has increased, and women are currently exposed to higher light-intensity levels.
The main limitation of the present study is that we Chronobiol Int Downloaded from informahealthcare.com by 85.250.145.131 on 12/23/10 Lastly, in the past decade, light bulbs emitting bluer were unable to measure precise light levels in the “sleep- light waves (∼460 nm) have been widely introduced to ing habitat,” which would have been a more accurate way save energy consumption and reduce CO2 emission.
of measuring LAN in the bedroom compared to the limits The results of a study carried out by Cajochen and col- of a questionnaire-based method. We attempted to leagues (2005) revealed that short-wavelength light compensate for this limitation, at least in part, by (460 nm) given for 2 h in the late evening decreased the application of a well-designed questionnaire and MLT production while increasing alertness, body temp- through study of a relatively large number of participants erature, and heart rate. Concurrently, exposure to a wave- (n = 1679), thereby resulting in good statistical power.
length of 550 nm for the same duration of time did not Nevertheless, more precise measurements of light levels in the “sleeping habitat” should be investigated Better understanding in the last decades of the way by in follow-up studies. Moreover, the fact that the response which light is transferred from the retina to different rate of the invited participants was not 100% may also be areas of the brain has opened new areas for research.
viewed as an additional limitation of the study.
The discovery of the non–image-forming photoreceptors Beyond doubts, LAN is an integral part of our lifestyle (NIFPs) and the non-vision pigment melanopsin (Berson and that in the future LAN levels may increase. Fortu- et al., 2002) enhanced chronobiological research, showing nately, this is an environmental variable that can easily that even a subterranean rodent, such as the “blind” be controlled, thus reducing its health risk. Simple sol- mole rat Spalax ehernbergi, is photosensitive to different utions exist to avoid the penetration of street illumination into the “sleeping habitat,” such as by the installation of Kloog I, Haim A, Stevens RG, Barchana M, Portnov BA. (2008). Light at window blinds in the bedroom, and also by the use of night co-distributes with incident breast but not lung cancer in the female population of Israel. Chronobiol. Int. 25:65–81.
Kloog I, Haim A, Portnov BA. (2009a). Using kernel density function as an urban analysis tool: Investigating the association between night- light exposure and the incidence of breast cancer in Haifa, Israel.
Comput. Environ. Urban Sys. 33:55–63.
We are grateful to Prof. Gad Rennert of the Carmel Kloog I, Haim A, Stevens RG, Portnov BA. (2009b). Global co-distri- Medical Center and Bruce Rappaport, the Faculty of bution of light at night (LAN) and cancers of prostate, colon, andlung in men. Chronobiol. Int. 26:108–125.
Medicine of Technion, for incorporating LAN-related Kloog I, Stevens RG, Haim A, Portnov BA. (2010, 3 August). Nighttime questions to an ongoing case-control study of the mol- light level co-distributes with breast cancer incidence worldwide.
ecular and environmental etiology of breast cancer in Cancer Causes Control [ePub ahead print], DOI: 10.1007/s10552- Israel and granting us access to the survey results. Our gratitude is also due to three anonymous reviewers for Nelson RJ. (2004). Seasonal immune function and sickness responses.
their numerous helpful comments and suggestions.
O’Leary ES, Schoenfeld ER, Stevens RG, Kabat GC, Henderson K, Grimson R, Gammon MD, Leske MC. (2006). Shift work, light at night, and breast cancer on Long Island, New York.
of interest. The authors alone are responsible for the Pauley SM. (2004). Lighting for the human circadian clock: recent re- search indicates that lighting has become a public health issue.
Med. Hypotheses 63:588–596.
Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods for biologic research on animals and human beings. Chronobiol.
Berson DM, Dunn FA,Takao M. (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295:1070–1073.
Rennert G, Pinchev M, Rennert HS. (2010). Use of bisphosphonates Blask DE, Brainard GC, Dauchy RT, Hanifin JP, Davidson LK, Krause and risk of postmenopausal breast cancer. J. Clin. Oncol.
JA, Sauer LA, Rivera-Bermudez MA, Dubocovich ML, Jasser SA, Lynch DT, Rollag MD, Zalatan F. (2005). Melatonin-depleted Rothman K, Greenland S. Lash T. (2008). Modern epidemiology. Balti- blood from premenopausal women exposed to light at night stimu- more, MD: Lippincott Williams & Wilkins.
lates growth of human breast cancer xenografts in nude rats. Cancer Schernhammer ES, Laden F, Speizer FE, Willett WC, Hunter DJ, Kawachi I, Colditz GA. (2001). Rotating night shifts and risk of Cajochen C, Munch M, Kobialka S, Krauchi K, Steiner R, Oelhafen P, breast cancer in women participating in the nurses’ health study.
Orgul S, Wirz-Justice A. (2005). High sensitivity of human melato- J. Natl. Cancer Inst. 93:1563–1568.
nin, alertness, thermoregulation, and heart rate to short wave- Schernhammer ES, Kroenke CH., Laden F, Hankinson SE. (2006).
length light. J. Clin. Endocrinol. Metab. 90:1311–1316.
Night work and risk of breast cancer. Epidemiology 17:108–111.
Damianaki A, Bakogeorgou E, Kampa M, Notas G, Hatzoglou A, Pana- Soleas GJ, Grass J, Josephy PD, Goldberg DM, Diamandis SP. (2002). A giotou S, Gemetzi C, Kouroumalis E, Martin PM Castanas E. (2000).
Potent inhibitory action of red wine polyphenols on human breast comparison of the anticarcinogenic properties of four red wine cancer cells. J.Cell Biochem. 78:429–441.
polyphenols. Clin. Biochem. 35:119–124.
Davis S, Mirick DK, Stevens RG. (2001). Night shift work, light at night, Srinivasan V, Spence D, Pandi-Perumal S, Trakht I, Esquifino A, Cardi- and risk of breast cancer. J. Natl Cancer Inst. 93:1557–1562.
nali D, Maestroni G. (2008). Melatonin, environmental light, and Feychting M, Osterlund B,Ahlbom A. (1998). Reduced cancer inci- breast cancer. Breast Cancer Res. Treat. 108:339–350.
dence among the blind. Epidemiology 9:490–494.
Stevens RG, Blask DE, Brainard GC, Hansen J, Lockley SW, Provencio I, Haim A, Heth G, Pratt H, Nevo E. (1983). Photoperiodic effects on ther- Rea MS, Reinlib L. (2007). Meeting report: the role of environ- moregulation in a ‘blind’ subterranean mammal. J. Exp. Biol.
mental lighting and circadian disruption in cancer and other dis- eases. Environ. Health Perspect. 115:1357–1362.
Chronobiol Int Downloaded from informahealthcare.com by 85.250.145.131 on 12/23/10 Haim A, Shanas U, Zubidad AS, Scantelbry M. (2005). Seasonality and Straif K, R. Baan R, Grosse Y, Secretan BE, El-Ghissassi F, Bouvard V, seasons out of time—the thermoregulatory effects of light interfer- Altieri A, Benbrahim-Tallaa L, Cogliano V. (2007). Carcinogenicity of shift-work, painting, and fire-fighting. Lancet Oncol. 1 Hansen J. (2001). Light at night, shiftwork, and breast cancer risk. J.
Vaughan TA. (1972). Mammalogy. Philadelphia: WB Saunders.
Haus E. (2009). Chronobiology in oncology. Int. J. Radiat. Oncol. Biol.
Verkasalo PK, Pukkala E, Stevens RG, Ojamo M, Rudanko SL. (1999).
Inverse association between breast cancer incidence and degree ICBS. (2010). Israel Central Bureau of Statistics, Jerusalem (http:// of visual impairment in Finland. Br. J. Cancer 80:1459–1460.
Zubidat AE, Nelson RJ, Haim A. (2009). Photosensitivity to different Kesteloot H. (2004). Alcohol intake and markers of inflammation. Eur.
light intensities in blind and sighted rodents. J. Exp. Biol.
Kliukiene J, Tynes T, Andersen A. (2001). Risk of breast cancer among Zubidat AE, Nelson RJ, Haim A. (2010). Photoentrainment in blind and Norwegian women with visual impairment. Br. J. Cancer sighted rodent species: responses to photophase light with different

Source: http://lichthinder.preventielichthinder.be/sites/all/documenten/CBI%20Urban%20Light.pdf