Intervenor: vol. 27, no. 1- 2, January - June 2002

Children's Health Project: Update on Childhood Cancer

Since publishing Environmental Standard Setting and Children's Health in May of 2000, we have updated our information on childhood cancer trends and also found an error in Chapter 2 of the report. An erratum and clarification has been posted to the CELA website. Anyone in possession of the report should download the erratum so that it can accompany the report.

This article includes and expands upon the information provided in the erratum and clarification.

In Chapter 2 of Environmental Standard Setting and Children's Health, entitled "Relationship Between Children's Health and Environmental Contaminants," in Section 2.6.9, page 70-71 the following two statements are made:

Although the number of children affected is small, there is evidence of increased incidence of childhood cancers. In Canada, there has been a 25% increase in the last 25 years in cancer incidence among children under 15 years of age.

The sources cited for these statements were the following:

Canadian Institute of Child Health (CICH). What on Earth? Proceedings from National Symposium on Environmental Contaminants and the Implications for Child Health - Canadian Institute for Child Health (May 1997); and National Cancer Institute of Canada. Canadian Cancer Statistics (Toronto, Canada. 1995).

The proceedings for Session 6 (Childhood Cancer) of the above-named CICH symposium provided a summary of a presentation by Dr. Howard Morrison of the Cancer Bureau, Health Canada who was cited as stating that childhood cancer incidence had increased in Canada. The National Cancer Institute of Canada reference should have been for the year 1997 and not 1995. The 1997 National Cancer Institute of Canada source reports (in a graphical representation) that Canadian trends in childhood cancer incidence appear to have shown a small but steady increase over the past 30 years.

Unfortunately, the second statement about the magnitude of any presumed increase in Canadian childhood cancer was incorrectly cited. At the time this chapter in Environmental Standard Setting and Children's Health was researched and written (i.e., during 1998 and 1999) several U.S. sources described increases in the incidence of childhood cancer in that country. Available data from the Surveillance, Epidemiology, and End Results (SEER) Program of the U.S. National Cancer Institute (NCI) suggested that the incidence rates for all types of childhood cancer in the United States had shown on average an approximately 1% per year increase from 1974 to 1991.[1] This widely cited statistic from U.S. data provided the basis for an extrapolation of a 25% increased incidence over the last 25 years. The error in Environmental Standard Setting and Children's Health was to apply the original characterization of an average 1% per year increase in U.S. childhood cancer incidence to the Canadian situation. While it might be the case that Canadian incidence rate trends are comparable to those in the United States and other industrialized countries, there has as yet been no published analysis of long-term historical Canadian data that would support this statement.[2]

There is an ongoing debate among researchers as to whether the increasing incidence of childhood cancer anywhere is due to anything other than earlier and improved diagnosis. Some experts consider diagnostic improvements as sufficient explanation for the increases found in cancer incidence while others consider that additional factors, including environmental contaminants, may be involved. It is clear that analysis of temporal trends is challenging and that conclusions are different depending on the data sample analysed, the time frame compared and the analytical techniques used, among other factors. For example, different research groups had noted in the mid 1990s that the increased childhood cancer incidence in their study samples was largely explained by increasing incidence of central nervous system (CNS) tumours and by higher increases in very young children for certain types of cancer.[3]

Smith and colleagues (1998) of the U.S. NCI first explored the idea that increases in brain tumour incidence might reflect improvements in detection and reporting of brain tumours due to the widespread adoption of magnetic resonance imaging (MRI) techniques in the mid 1980s. Their study determined that using a "jump" model, rather than a linear model which averaged out yearly rate changes, better explained the data patterns and indicated that increases in brain tumour incidence rates reflected a sharp jump from 1984 to 1985 rather than a slow, steady increase over the whole period (1973 to 1994). It should be noted that further evaluation of the SEER data on all childhood cancers diagnosed between 1975 through 1995 by U.S. NCI researchers (headed by Martha Linet) concluded that the modestly increasing incidence was a temporary phenomenon again, likely reflecting diagnostic improvements or reporting changes, and that the incidence rates had stabilized (i.e., had not substantially increased or decreased) since the mid-1980s.[4,5]

In the same year as the study by Linet and colleagues, a report by Joseph Mangano presented an analysis of data from the nine registries in the SEER program plus data from two other states and one city.[6] His analysis focused on children under ten diagnosed from 1980 to 1993, comparing rates for an early and a later cohort. He determined that there had been a 37% increase in cancer rates between those two cohorts, suggesting that one possible factor may be fetal and infant exposure to low-dose radioactivity. Although Mangano's data set was larger than that used by Linet and colleagues, it has been suggested that because of differences in the sample and analytical techniques, his conclusions may not be warranted.[7]

A 2001 study reported in the journal Cancer reviewed data on incidence of solid tumours in over 2,500 children diagnosed from 1954-1998 in Northwest England. McNally and colleagues found that the increases in incidence in their sample were likely to be real, and not artifactual or explained only by diagnostic changes.[8] Although these researchers note that U.S. and U.K. trends and data may not be directly comparable, they suggest, "perhaps the U.S data should not be dismissed entirely as artifact".[9]

Noted expert in pediatric environmental health, Dr. Philip Landrigan from the Mount Sinai Medical Center in New York City said in an interview with Science Magazine, "I'm a pediatrician. I see children with brain cancer. It's inconceivable to me to imagine that 25 years ago we were missing one-third of children with this disease."[10]

Thankfully, as noted in our report, childhood cancer is a relatively rare condition and survival has improved for some (such as leukemia), but not all childhood cancers due to advancements in treatment. Yet cancer is still the most common cause of illness-related death in Canadian children over one year of age.[11] Consequently there is ample reason to continue to evaluate factors (including environmental exposures) that may increase the risk of cancer in the young and to take appropriate steps to reduce potentially carcinogenic exposures. Researchers point out that we continue to have very limited understanding of the precise causes of different cancers in children. Preventing cancer in children is a high priority on the basis of potential years of life lost, quality of life issues and the potential for long-term (late) health effects in survivors.[12]
In Canada, the overall picture for childhood cancer, for the time periods 1992-1996 and 1993-1997, is summarized as follows:

cancer was diagnosed in an average of 1266 (Canadian) children (aged 0-19) every year, and 249 died each year from the disease. Leukemia accounted for 26% of new cases and 32% of deaths due to cancer in children, and remains the most common of the childhood cancers. Cancers of the brain and spinal cord, the second most common group of childhood cancers, constituted approximately 17% of new cases and 24% of deaths, and lymphomas accounted for 16% of new cases and 8% of deaths.[13]

The most recent report on Canadian cancer[14] released April 18, 2002, looks at childhood cancer between 1994 and 1998 describing very similar statistics to those from previous reports. The annual "Special Topic" for the year 2002 addresses cancer incidence in young adults (aged 20-44 years) finding that cancer incidence increased slightly from 1987 to 1996 with important increases in several types of cancer. Rates increased substantially more among young women than young men, primarily due to excess cancers of the reproductive system. Significant increases were also found for non-Hodgkin's lymphoma and thyroid cancer in both sexes, for lung and brain cancer in women, and testis cancer in men. The latter agrees with the findings of Weir and colleagues (discussed in our 2000 report) indicating substantial increases in testicular cancer incidence particularly among adolescent and young adult males.[15] Cancer in the young adult age group is also rare but such increases are worrisome and are not yet explained. Given that cancer in young adults also reflects a relatively short latency compared to cancer at older ages, contributing factors could well have occurred during childhood. This further strengthens the need for research to investigate potential risk factors and to prevent exposures from as early an age as possible.

Additional discussion of the evaluation of environmental contaminants and childhood cancer can be found in a recent issue of Pediatric Child Health.[16] See also the Children's Environmental Health Project on the web site of the Canadian Association of Physicians for the Environment.[17]

1. Gurney JG, et al. Trends in cancer incidence among children in the U.S. Cancer 1996;78:532-41. The SEER program provides data from nine cancer registries and represents about 10% of the U.S. cancer cases. The Gurney et al. study had a sample of over 12,000 cases of childhood cancer diagnosed between 1974 and 1991.

2. Because of our smaller population base there are absolutely fewer childhood cancer cases in Canada compared to the United States. One source states that "reliable Canadian statistics comparing trends in children's cancers over decades are not available, according to Cancer Care Ontario" (Everyday Carcinogens: Stopping Cancer Before it Starts. Background Paper for Workshop on Primary Cancer Prevention. March 26-27, 1999, McMaster University, Hamilton, Canada. See endnote 6 therein. Available at: Therefore while comparisons can be misleading, it is worth noting that the most common forms of cancer in Canadian children are also the most common forms of cancer in the United States and that when translated into rates (numbers of cases per million children) Canadian and U.S. childhood cancer rates are similar. [Carozza SE et al. Childhood cancer incidence in North America, 1988-1992. In: Cancer in North America 1989-1993. Howe HL (Ed.) Sacramento, CA: North American Association of Central Cancer Registries. April 1997 Vol V-1-13. Available at:

3. Gurney et al., 1996 see note 1; Bunin GR. et al. Increasing incidence of childhood cancer: report of 20 years experience from the greater Delaware Valley Pediatric Tumor Registry. Paediatr Perinat Epidemiol 1996;10:319-38; Kenney LB. et al. Increased incidence of cancer in infants in the U.S.: 1980-1990. Cancer 1998;82:1396-400.

4. Linet MS, et al. Cancer surveillance series: recent trends in childhood cancer incidence and mortality in the United States, J. Natl Cancer Inst 1999;91:1051-8. This analysis included data on cancers in over 14,000 children under 15 years diagnosed from 1975 to 1995.

5. For a summary of the US statistics on cancer incidence and trends, see Fact Sheet 6.40, National Cancer Institute Research on Childhood Cancers at:

6. Mangano JJ. A rise in the incidence of childhood cancer in the United States. Int J Health Serv 1999;29(2):393-408.

7. See Bukowski JA. Critical assessment of opposing views on trends in childhood cancer. Int J Health Serv 2000;30:373-7 and Mangano JJ. Response to critique by John Bukowski. Int J Health Serv 2000;30:379-86.

8. McNally RJ, et al., Temporal increases in the incidence of childhood solid tumors seen in Northwest England (1954 - 1998) are likely to be real, Cancer, 2001 Oct 1;92(7):1967-76.

9. McNally et al. 2001, note 8 above. Page 1975.

10. Kaiser, J, Epidemiology: No Meeting of Minds on Childhood Cancer, Science Magazine, 1999 Dec 3: 286(5446): 1832-1834.) Note that the above citations and statements refer to cancer incidence data in England and the United States.

11. Julie Charbonneau, Environment Canada and Anthony Myres, Health Canada, Children's Environmental Health -The Canadian Experience. Presentation to the Joint Meeting of the Public Advisory Committee (JPAC) and the Expert Advisory Board on Children's Health and the Environment in North America, March 7, 2002, Mexico City. See also: Health Canada. Cancer Incidence. In: Measuring Up: A Health Surveillance Update on Canadian Children and Youth. Rusen ID, McCourt C. (Eds.) 1999.

12. Savitz, DA. Environmental exposures and childhood cancers: Our best may not be good enough. Am J. Pub Health 2001;Apr 91(4):562-3.; Gurney, JG. Brain cancer incidence in children: Time to look beyond the trends. Med Ped Oncol 1999;33:110-12.

13. National Cancer Institute of Canada: Canadian Cancer Statistics 2001, Toronto, Canada, 2001, (on-line at

14. National Cancer Institute of Canada: Canadian Cancer Statistics 2002, Toronto, Canada, 2002. Available at: and

15. Weir, H.K. , L.D. Marrett and V. Moravan. Trends in incidence of testicular germ cell cancer in Ontario by histologic subgroup, 1964-1996. CMAJ. 1999;160:201-205.

16. Chance, GW, Environmental contaminants and children's health: Cause for concern, time for action. Paediatr Child Health. 2001;Dec. 6(10):731-743.

17. and CAPE's Children's Environmental Health Project


Loren Vanderlinden is co-author of the report, Environmental Standard-Setting and Children's Health. Kathleen Cooper is a CELA Researcher and co-author of the same report.