Men working at night had approximately double the risk of those who did not work the night shift for many different cancers, according to a Canadian case–control study published in November 1 issue of the American Journal of Epidemiology.

"The results suggest that night work may increase cancer risk at several sites among men," say the researchers, led by Marie-Élise Parent, PhD, from the epidemiology and biostatistics unit at the University of Quebec in Laval, Canada.

"The observation here of elevated risks for several other types of cancer is novel," they write. If our findings are valid, it would signal an important systemic cancer hazard," they add.

The study compared 3137 men with cancer and 512 matched control subjects who completed a detailed questionnaire about occupational work and lifestyle from 1979 to 1985.

Compared with men who never worked at night, men who worked night shifts had an increased risk for prostate cancer (odds ratio [OR], 2.77), non-Hodgkin's lymphoma (OR, 2.31), pancreatic cancer (OR, 2.27), rectal cancer (OR, 2.09), colon cancer (OR, 2.03), bladder cancer (OR, 1.74), and lung cancer (OR, 1.76).

"This is a top-notch study in a top journal," said Richard Stevens, PhD, a cancer epidemiologist at the University of Connecticut Health Center in Farmington. He was not involved in the study, but acted as a reviewer for the paper. "This is new" because the finding is in men and so many types of cancer are involved, he said.

The finding that so many different types of cancer are involved is surprising but plausible, he noted. "When I first proposed the light-at-night hypothesis, it linked circadian disturbance to hormone disruption and breast cancer," Dr. Stevens explained. Since then, circadian disruption has been linked to many other adverse effects, it has been demonstrated that melatonin has a direct effect on tumor growth, and some of the latest research he has been involved with has shown that circadian disruption has an epigenetic impact.

No Concern About Bias

Dr. Stevens said this is a high-quality study, and he is inclined to believe that the increase in cancer risk associated with night-shift work is "real."

There are 2 main limitations to case–control studies, he explained: recall bias and sampling bias. Neither is a concern in this study, he noted.

Recall bias related to ongoing publicity about night-shift work does not apply, because the field work was carried out in the 1980s, when the link between night-shift work and cancer was not clear, Dr. Stevens explained. Originally, the researchers were looking for anything in the working environment that would increase the risk for cancer; they have published studies on the association between occupational exposure to lead (Am J Epidemiol. 2007;166:1005-1014) and diesel/gasoline (Am J Epidemiol. 2007;165:53-62) and the risk for cancer.

Sampling bias comes into play when the control population is not representative of the population at large, which can skew the findings. Generally, the "cases" (for example, cancer patients) are more willing to participate and answer a lengthy questionnaire than the healthy matched "controls."

The best scenario is 100% participation; any less raises the possibility of sampling bias, he said. In this study, 70% of the control subjects who were approached agreed to participate, which is "pretty good," Dr. Stevens noted. In fact, the proportion of subjects in the control group working the night shift is similar to that found in the general Canadian population, which suggests that this cohort is representative, and reduces the likelihood of bias, he explained.

The researchers themselves are cautious in their conclusions. "The consistent positive association could not be ascribed to an identifiable methodological bias, but chance cannot be ruled out as an explanation for these findings," they write.

"Our results lend some support to the hypothesis that night work might lead to an increase in cancer risk," they conclude.

"With accruing epidemiological and experimental evidence, and because of the increasing prevalence of night work, further exploration of the hypothesis is warranted," they add.

The evidence of an association is increasing, acknowledge Parveen Bhatti, PhD, and colleagues from the Fred Hutchinson Cancer Research Center in Seattle, Washington, in an accompanying commentary. They note that the International Agency for Research on Cancer recently classified night-shift work as "possibly carcinogenic." In response to that classification, in 2009, a Danish court awarded compensation to women with breast cancer who had worked night shifts for long periods of time.

Dr. Bhatti and colleagues also call for more research on this subject. "Epidemiological studies with detailed data on the various aspects of shift work are needed to truly understand the associations between shift work and multiple types of cancer that are being increasingly observed," they conclude. These could take years, however. In the meantime, studies incorporating biomarkers of the effect of night-shift work might be a useful alternative, they add.

Am J Epidemiol. 2012;176:751-759, 760-763. Abstract, Commentary