Journal of Public Health Advance Access originally published online on April 28, 2008
Journal of Public Health 2008 30(2):194-201; doi:10.1093/pubmed/fdn019
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gender differences in colorectal cancer incidence, mortality, hospitalizations and surgical procedures in Canada
Ru-Nie Gao, Senior Epidemiologist1
C. Ineke Neutel, Senior Epidemiologist1,2
Elaine Wai, Radiation Oncologist3
1 Surveillance Division, Centre for Chronic Disease Prevention and Control, Public Health Agency of Canada, 785 Carling Avenue, Ottawa, Canada
2 Department of Epidemiology and Community Medicine, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Ottawa, Ontario, Canada K1H 8M5
3 Department of Radiation Oncology, BC Cancer Agency – Vancouver Island Centre, 600 West 10th Avenue, Vancouver, BC, Canada V5Z 4E6
Address correspondence to Dr Ru-Nie Gao, E-mail: ru-nie_gao{at}phac.aspc.gc.ca
Background Over the past few decades there have been changes in incidence and mortality of colorectal cancer.
Objective To examine gender differences in incidence, hospitalization, hospital-based procedures and mortality for colorectal cancer.
Methods Data were derived from the Hospital Morbidity Database, Canadian Cancer Registry and the Canadian Mortality Database.
Results Overall incidence and mortality rates for colorectal cancer are decreasing, but remain substantially higher for males. Absolute numbers of cases are similar for men and women. The top subsite for men was rectal cancer, which was third highest for women, whereas right colon cancer was highest for women. Male/female ratios for incidence and surgeries were highest for distal cancer and are increasing with time.
Conclusions Although overall incidence rates have shown a decline, absolute numbers of new colorectal cancer cases have increased. While men have higher colorectal cancer rates, women have similar numbers and screening should target both equally. Over the years, colorectal cancer subsites are showing a rightward shift, i.e. an increase in proximal subsites, but a leftward shift in male/female ratios, i.e. a greater decrease for the more distal subsites in females. The lower rates for women for distal cancer are compatible with a degree of hormonal protection based on oral contraceptive and hormone replacement therapy. Colorectal cancer will continue to be a considerable public health problem in the foreseeable future.
Keywords: colorectal cancer, incidence, mortality, surgical procedures, survival, trends
 |
Introduction |
|---|
 TOP Introduction MethodologyResultsDiscussionConclusionsReferences |
|---|
Colorectal cancer remains a serious problem. In fact, in Canada, it is the third most common cancer in terms of incidence rates with only lung cancer contributing more cancer deaths.1–3 As many as 1 in 16 Canadian men and 1 in18 Canadian women are expected to develop colorectal cancer in their lifetime.2 Although age-adjusted incidence rates for males have stayed relatively constant over the last few decades, incidence rates for females, already lower than those of men, have been declining. The result is an increasing ratio of male to female age-adjusted incidence rates.1,2 These ratios were found to be the lowest, i.e. nearly 1, for proximal colon cancer cases, higher for distal colon cancer and highest for rectal cancer at a ratio of
2.2,4 The age-adjusted mortality rates for males did not start decreasing until about 1990, while females showed a lower mortality to start with and started to decline earlier.1–3 Colorectal cancer is very much a cancer of the elderly and with increasing aging of the population the number of cases in the population may be expected to rise further in spite of decreases in incidence and mortality rates.1,2 Thus, it is important to remain vigilant regarding the changing pattern of colorectal cancer. The objective of this article is to examine trends in colorectal cancer incidence, mortality, hospitalizations and in-hospital surgical procedure rates over the last few decades. A special emphasis in this article will be on the changing gender differences in colorectal cancer.3 |
Methodology |
|---|
|
TOPIntroductionMethodologyResultsDiscussionConclusionsReferences |
|---|
Data were obtained from the Hospital Morbidity Database, the Canadian Cancer Registry, and the Canadian Mortality Database, all of which are collected by the Canadian provinces and territories, after which the data are compiled and maintained nationally by Statistics Canada. The Hospital Morbidity Database supplied data on all in-patient hospitalizations with a primary diagnosis of colorectal cancer for the fiscal years 1981–82 to 2000–01. Although the fiscal years include parts of two adjacent years, for the sake of simplicity, we will refer to the fiscal year by the first year, e.g. the fiscal year of 1 April 1981 to 31 March 1982 will be referred to as 1981.
Hospital data are event-oriented, so that for persons with multiple hospital stays each hospital stay is treated as a separate event. Colorectal cancer hospital stays were selected based on a primary diagnosis of colorectal cancer as codes 153, 154 in the International Classification of Diseases version nine (ICD-9).5 Surgical procedures were coded as in the Canadian Classification of Diagnostic and Therapeutic Procedures (CCP).6 Procedures used in the subsite analysis were right hemicolectomy (57.53), transverse resection (57.54), left hemicolectomy (57.55), sigmoidectomy (57.56), abdominoperineal resection of the rectum (60.4) which includes a permanent colostomy, and anterior resection of the rectum (60.5) which does not usually include a permanent colostomy. If more than one type of surgery was mentioned, the procedure for that hospital stay was coded to the more proximal cancer mentioned. Hospital procedures were also categorized as ‘other procedures’, i.e. a procedure other than those already mentioned, or ‘no procedure’ if the procedure code was left blank. Hospitalizations for cecectomy (57.22), other partial excision (57.59), total colectomy (57.6), ‘other procedures’ and ‘no procedure’ were included in the analysis for all colorectal hospitalizations combined, but not in the trend analysis of major surgical procedures.
The Canadian Cancer Registry was used to identify colorectal cancer incidence data by codes 153, 154 ICD-9 for the years 1981–91 and ICO-0-3 as C18-21 for 1992–2001.7 The Canadian Mortality Database (1981–2001) was used to identify colorectal cancer deaths by an underlying cause of death coded ICD-9, codes 153 and 154. The analysis of nationwide data consisted of calculated age-specific rates using the Canadian population based on Statistics Canada census data as denominator. Age-adjusted rates were calculated using Canadian population of 1991 as the standard population. Moving averages were calculated using the MS Excel ‘moving average’ feature, which inserts the average of three points in the place of the third of the original values. Thus, curves start in 1983 even though the data start in 1981.
|
Results |
|---|
|
TOPIntroductionMethodologyResultsDiscussionConclusionsReferences |
|---|
Over the years 1981–2001, 313 350 new cases of colorectal cancer were diagnosed in Canada, 417 223 cases were hospitalized and 147 447 died (Table 1). Note that when the data were grouped into four year periods, one year was left out and thus the totals for the years categories are lower than those for the age or sex categories. The highest age-adjusted rates for incidence, mortality and hospitalization were in the oldest age group,
80 years, although the largest number of new cases and hospitalizations occurred in the 60–69- and 70–79-year age groups. The numbers of new cases, hospitalizations and deaths continued to increase over the years in spite of decreasing rates.
|
All age-adjusted incidence, mortality and hospitalization rates for colorectal cancer were substantially higher for males than for females (Fig. 1a and b). Hospitalization rates showed declining trends for both genders until 1994 after which rates leveled out. Incidence and mortality rates showed small decreases for both males and females. In spite of the decreasing rates, absolute numbers of new cases, deaths and hospital stays continued to increase with numbers for males and females more similar to each other than were the rates (Fig. 1c and d).
|
Trends in age-adjusted incidence rates by subsite show considerable differences by gender (Fig. 2a and b). Rectal cancer (for this table excluding rectosigmoid cancer) rates accounted for the highest incidence rates for males, but only third for women. The site with the highest incidence for women was right colon, even though this rate was still lower than that for right colon cancer in men and increased less. Other differences in trends were that the sigmoid cancer incidence rate declined for women but remained nearly the same for men, and rectosigmoid cancer incidence increased for males but remained same for females. The numbers of new cases rose for both men and women for all subsites even where age-adjusted rates decreased (Fig. 2c and d).
|
Right hemicolectomy was the most frequently performed surgical procedure for males and females although the rate of anterior resection for males increased at such a rate that it nearly equaled right hemicolectomy by the year 2001 (Fig. 3a and b). Anterior resection rates for women started lower and increased less. The rates for abdominoperineal resection and anterior resection, the two main surgeries for rectal cancer, showed quite different trends. While the rate of anterior resection increased dramatically, the rate of abdominoperineal resection decreased moderately. Transverse colon resection was the only procedure for which rates and trends were similar for males and females. In terms of absolute number of hospitalizations, right hemicolectomies were by far the most frequently performed procedure with the numbers even higher for women than for men. Anterior resection was next, although here absolute numbers were much lower for women (Fig. 3c and d). The frequency for these two surgeries continued to increase.
|
The male/female ratios for right colon cancer incidence and right hemicolectomy rates, transverse colon cancer incidence and transverse colon resection, indicate that time trends were the same for men and women (Table 2). Male/female ratios for left colon, sigmoid colon and rectal cancer age-adjusted incidence, as well the corresponding surgeries, left hemicolectomy and resection of the sigmoid colon and rectum, respectively, showed increasing difference between males and females, in spite of differences in rates between the age-adjusted incidence and the corresponding surgical procedure rates.
|
|
Discussion |
|---|
|
TOPIntroductionMethodologyResultsDiscussionConclusionsReferences |
|---|
Main findings
This study has identified different trends in incidence and mortality rates over the years with a shift in location of incidence rates of colorectal cancer. Overall incidence and mortality rates for colorectal cancer showed decreasing trends over the years especially for females, and both continue to be substantially higher for males than for females. In spite of the lower rates for women, the absolute number of cases for the two sexes is more similar than one might expect. This is due to the fact that colorectal cancer is most common for the older age groups in which women outnumber men.
A surprising finding was that males and females showed such differences in the distribution of subsites. For example, males had the highest rates for rectal cancer, which was only the third highest for women at just above half the male rates. The most common subsite for females was right colon cancer, still somewhat lower than the male equivalent. Subsite-specific incidence rates for females showed declines or remained level with the possible exception of right colon cancer, which showed a slight increase in recent years. Subsite-specific incidence rates for males stayed level or showed increases, giving the impression of a leftward shift in the differences between men and women. Given the rightward shift in colorectal cancer sites for both men and women, and the leftward shift in differences between men and women, it is clear that these changes in colorectal cancer rates have a public health significance which will require constant vigilance and repeated examination and monitoring in trends of colorectal cancer and its subsites.
Strengths and limitations
Important strengths are the complete and consistent information on incidence, hospitalization, hospital procedures and mortality for the entire Canadian population over the two decades. Relevant limitations would include the lack of desirable information such as data on cancer staging and outpatient procedures or treatment, such as radiotherapy or chemotherapy. For many cancers, the lack of information on day surgery would be a concern, but few colorectal cancer surgeries are done on a day surgery basis.8
Another issue is the event-oriented nature of hospital data with the possibility of more than one record per person if that person had more than one hospital stay. Both event- and person-oriented data have their uses. From the perspective of health resource utilization, event-oriented data are useful, whereas from the perspective of disease etiology or disease management in individual patients, patient-oriented data are preferable. The difficulty with event-oriented data in this study is in comparability between incidence and hospital data when one set is person oriented and the other is event oriented. However, since our interest is in trends rather than in isolated absolute numbers or precise rates, we can still draw conclusions because of the consistency of the measurement over the years of the study.
Even with these limitations, it is paramount that we examine the rates produced on a regular basis since these data are available for the whole country on an ongoing basis. In general, the large amounts of data contained in these databases are invaluable in surveillance and need to be monitored extensively before more expensive and time consuming ad hoc studies are done.
What this study adds to what is already known
Putting the findings of this study in the context of what is already known, we can start with noting that the modest decreases in age-adjusted mortality rates for both men and women, and in incidence rates especially for women, are encouraging. Such declines are consistent with some other studies,9,10 but not all.11 In spite of these decreasing trends in age-adjusted mortality, incidence and hospitalization rates, the absolute numbers of new cases, hospitalizations and deaths are still increasing, largely because of the aging of the population. Consequently, colorectal cancer still has a large impact on health care services, and this impact will continue to increase for many more years.9,12 Studies like the present one are a warning that health care services need to be prepared.
Interesting, although somewhat puzzling, findings are the shifts in subsites. The age-adjusted incidence rates by subsite in this study confirm what elsewhere has been called a ‘rightward shift’ in colorectal cancer subsite distribution.8,11,13 In this study, right colon cancer incidence was shown to increase, especially for men, whereas transverse colon and sigmoid colon cancers remained same for both males and females, and rectal and sigmoid cancer incidences, especially for men, decreased. US studies also found a rightward shift but found that this was mostly due to reduction in left-sided colon cancer while the proportions of right-sided cancer stayed relatively similar, rather than the increase in right-sided cancer as seen in Canada.13–15
Rabeneck et al.16 suggested that the rightward shift may be explained by the aging of the population since right-sided cancer tends to occur more often in the older people. However, the age-adjusted rates used in the present study should have compensated for the effects of aging. Gomez et al.17 attributed the rightward shift to improved diagnosis of more proximal cancers by means of the longer range of modern flexible sigmoidoscopes. This would not explain the gradual change nor the increasing differences between men and women. No satisfactory reason can yet be advanced to explain the reason for the rightward shift of colorectal cancer sites. More work needs to be done in this area since explaining the shift may well have a bearing on prevention.
Contrary to the overall rightward shift of colorectal cancer sites, differences between males and females are undergoing a leftward shift. This leftward shift manifests itself as increasing differences between men and women in the more distal sites, i.e. left colon, sigmoid and rectum, where differences were already the greatest. Thus, in the most recent time period, the incidence of rectal cancer in men is double that of women. It is hard to know why these differences exist and why the magnitude of the differences is still increasing. It is possible that there are physiological differences between men and women that lead to differences in incidence. For example, there is the possibility of a hormonal factor and several studies have shown that oral contraceptive (OC) use and hormone replacement therapy (HRT) have a protective effect for both colon and rectal cancer.18–20 Considering the increasing use of OC and HRT decades ago and assuming a long latent period in developing or preventing cancer, the protective effects of hormonal use may have been reducing female rectal cancer in more recent decades, leading to increasing differences between males and females. Possibly, the recent decrease in HRT use may eventually halt the increasing difference between men and women.
Age-adjusted rates of hospital surgical procedures rates show changes corresponding to the already mentioned changes in site-specific incidence. Thus, both males and females show increases in rates of right hemicolectomy corresponding to the changes in the incidence of right colon cancer, as well as trends in the rates of left hemicolectomy corresponding to the changes in left colon cancer incidence. The correspondence between the male/female ratios based on site-specific incidence and those based on hospital procedures confirm the value of these ratios, especially considering the differences in the measurement of these rates. The possibility of multiple hospitalizations per person may explain part of the difference between incidence and hospitalization rates for some sites. Another possible difference is that there is not always an exact anatomical correspondence between incidence and surgery sites. For example, right hemicolectomy is not necessarily confined to the exact anatomical site of right colon and some cancers assigned to transverse colon for incidence may well be treated by a right hemicolectomy procedure. In spite of these differences, the male/female ratios are surprisingly similar for site-specific incidence rates and the corresponding hospitalization rates, which strengthen confidence in the trends shown by the ratios.
A change that relates more to development in surgery rather than to changing anatomical sites is that in rectal surgery, rates of anterior resection are noticeably surpassing abdominoperineal resection. Most likely, patients and surgeons are preferentially choosing this procedure because it does not include a permanent colostomy.21 The greater decrease in incidence rates of rectal cancer for women may explain why anterior resection has increased even more for men than for women. Although anterior resection has increased greatly, the use of abdominoperineal surgery has decreased only a modest amount, perhaps less than one would expect. Since hospital data are event based, it is possible that there are multiple hospital procedures per cancer incidence case.
|
Conclusions |
|---|
|
TOPIntroductionMethodologyResultsDiscussionConclusionsReferences |
|---|
The findings of this study add a renewed perspective on the considerable public health impact of colorectal cancer. Any modest decreases in incidence and mortality rates seen in this study are more than offset by the increasing numbers of new cases and here the increases show no sign of abating. It is clear that colorectal cancer continues to be a serious health problem. This will be magnified by the aging of the population unless new knowledge will lead to prevention. This study emphasizes the changes in the frequency of subsites occurring without any changes in the aggregate rates. The mere occurrence of these shifts suggests that there are environmental risk factors at play with the term ‘environmental’ interpreted in the broadest possible way. Learning more about the factors that affect these shifts in colorectal cancer may be a huge step toward learning new means of prevention.
Early detection remains the brightest hope for minimizing the impact of colorectal cancer on society. Colorectal cancer is a good target for screening because of the long pre-malignant course and the favorable response to early intervention.22,23 Annual screening for two-thirds of people between ages 50 and 74 is expected to reduce 10-year colorectal cancer mortality by 26%.24 Although few populations received adequate screening for colorectal cancer,25 women tended to receive screening tests even less often than men.26,27 It is possible that the lower rates of colorectal cancer in women may have led physicians to consider the risk of colorectal cancer for women as being less serious and thus early detection less important.28 Still, the number of new cases or number of hospitalizations are very similar for men and women. On the whole, more aggressive screening is needed for both men and women.
|
References |
|---|
|
TOPIntroductionMethodologyResultsDiscussionConclusionsReferences |
|---|
- National Cancer Institute of Canada. Canadian Cancer Statistics (2006) Toronto: Canada.
- Gibbons L, Waters C, Mao Y, et al. Trends in colorectal cancer incidence and mortality. Health Rep (2001) 12:41–55.[Medline]
- National Cancer Institute of Canada. Review of Trends in Colorectal Cancer (2003) http://www.ncic.cancer.ca/ncic/, (accessed March, 2008).
- Cheng X, Chen VW, Steele B, et al. Subsite-specific incidence rate and stage of disease in colorectal cancer by race, gender, and age group in the United States, 1992–1997. Cancer (2001) 92:2547–54.[CrossRef][Web of Science][Medline]
- World Health Organization. International Classification of Diseases (1977) 9th ed. Geneva: WHO.
- Statistics Canada. Canadian Classification of Diagnostic. Therapeutic and Surgical Procedures. (1993) Ministry of Industry; Catalogue No. 82-562E.
- Fritz A, Percy C, Jack A, et al. International Classification of Diseases for Oncology (ICD-O) (2000) Geneva: World Health Organization.
- Neutel CI, Gao RN, Wai E, Gaudette LA. Trends in in-patient hospital utilization and surgical procedures for breast, prostate, lung and colorectal cancers in Canada. Cancer Causes Control (2005) 16:1261–70.[CrossRef][Web of Science][Medline]
- Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer statistics, 2001. Cancer J Clin (2001) 51:15–36.
- Fernandez E, La Vecchia C, Gonzalez JR, et al. Converging patterns of colorectal cancer mortality in Europe. Eur J Cancer (2005) 41:430–7.[CrossRef][Web of Science][Medline]
- Centers for Disease Control and Prevention (CDC). Recent trends in mortality rates for four major cancers, by sex and race/ethnicity—United States, 1990–1998. MMWR Morb Mortal Wkly Rep (2002) 51:49–53.[Medline]
- Capocaccia R, De Angelis R, Frova L, et al. Estimation and projections of colorectal cancer trends in Italy. Int J Epidemiol (1997) 26:924–32.
[Abstract/Free Full Text] - Troisi RJ, Freedman AN, Devesa SS. Incidence of colorectal carcinoma in the U.S.: an update of trends by gender, race, age, subsite, and stage, 1975–1994. Cancer (1995) 85:1670–6.[CrossRef]
- Nelson RL, Persky V, Turyk M. Time trends in distal colorectal cancer subsite location related to age and how it affects choice of screening modality. J Surg Oncol (1998) 69:235–8.[CrossRef][Web of Science][Medline]
- Gupta AK, Melton LJ, Petersen GM, et al. Changing trends in the incidence, stage, survival, and screen-detection of colorectal cancer: a population-based study. Clin Gastroenterol Hepatol (2005) 3:150–8.[CrossRef][Web of Science][Medline]
- Rabeneck L, Davila JA, El-Serag HB. Is there a true "shift" to the right in the incidence of colorectal cancer? Am J Gastroenterol (2005) 98:1400–9.
- Gomez D, Dalal Z, Raw E, et al. Anatomical distribution of colorectal cancer over a 10 year period in a district general hospital: is there a true "rightward shift"? Postgrad Med J (2004) 80:667–9.
[Abstract/Free Full Text] - Fernandez E, La Vecchia C, Balducci A, et al. Oral contraceptives and colorectal cancer risk: a meta-analysis. Br J Cancer (2001) 84:722–7.[CrossRef][Web of Science][Medline]
- Kampman E, Potter JD, Slattery ML, et al. Hormone replacement therapy, reproductive history, and colon cancer: a multicenter, case-control study in the United States. Cancer Causes Control (1997) 8:146–58.[CrossRef][Web of Science][Medline]
- Nichols HB, Trentham-Dietz A, Hampton JM, et al. Oral contraceptive use, reproductive factors, and colorectal cancer risk: findings from Wisconsin. Cancer Epidemiol Biomarkers Prev (2005) 14:1212–8.
[Abstract/Free Full Text] - Semmens JB, Platell C, Threlfall TJ, et al. A population-based study of the incidence, mortality and outcomes in patients following surgery for colorectal cancer in Western Australia. Aust NZJ Surg (2000) 70:11–8.[CrossRef]
- Ciccolallo L, Capocaccia R, Coleman MP, et al. Survival differences between European and US patients with colorectal cancer: role of stage at diagnosis and surgery. Gut (2005) 54:268–73.
[Abstract/Free Full Text] - Martinez SR, Young SE, Hoedema RE, et al. Colorectal cancer screening and surveillance: current standards and future trends. Ann Surg Oncol (2006) 13:768–75.[CrossRef][Web of Science][Medline]
- Flanagan WM, Le Petit C, Berthelot JM, et al. Potential impact of population-based colorectal cancer screening in Canada. Chronic Dis Can (2003) 24:81–8.[Medline]
- Rabeneck L, Paszat LF. A population-based estimate of the extent of colorectal cancer screening in Ontario. Am J Gastroenterol (2004) 99:1141–4.[CrossRef][Web of Science][Medline]
- Meissner HI, Breen N, Klabunde CN, et al. Patterns of colorectal cancer screening uptake among men and women in the United States. Cancer Epidemiol Biomarkers Prev (2006) 15:389–94.
[Abstract/Free Full Text] - Slattery ML, Edwards SL, Ma KN, et al. Colon cancer screening, lifestyle, and risk of colon cancer. Cancer Causes Control (2000) 11:555–63.[CrossRef][Web of Science][Medline]
- Lieberman D. Race, gender, and colorectal screening. Am J Gastroenterol (2005) 100:2756–8.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||