Publications

    Health Research Working Paper Series

    The feasibility of adding treatment data to the Canadian Cancer Registry using record linkage

    Results

    Warning View the most recent version.

    Archived Content

    Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

    Cancer cohorts

    For 2005 through 2008, the following primary, single, malignant tumour cohorts were identified in the CCR: 37,984 female breast tumours; 36,171 colorectal tumours; and 43,448 prostate tumours (Table 1).

    Eligibility for linkage

    Table 2 displays the numbers and percentages of tumours eligible for linkage to the DAD and/or NACRS. Throughout the 2005-to-2008 period, linkage eligibility rates of all tumours in a given cancer cohort ranged from 97.0% to 100% in Manitoba; from 98.1% to 99.0% in Ontario; from 99.1% to 100% in Nova Scotia; and from 98.8% to 100% in Prince Edward Island. No systematic biases in eligibility were evident overall, or according to patient and tumour characteristics (data not shown).

    Linkage rates

    Table 3 displays linkage rates of tumours to the DAD, and for Ontario, to the NACRS, by province and hospital service type from 2005 to 2008. For breast cancer tumours, linkage to the DAD/NACRS was consistently above 95% in all provinces and years, except Manitoba in 2008 (94%).

    The percentages of breast tumours that linked to acute care hospitalizations ranged from 52% (Ontario 2008) to 96% (Prince Edward Island, 2006). In Nova Scotia, linkage rates dropped from about 83% (2005 and 2006) to 75% in 2007, and to 73% in 2008. Linkage rates of breast tumours to same-day surgery ranged from 63% (Ontario, 2005) to 75% (Nova Scotia, 2008); the range in Prince Edward Island was lower (34% to 51%).

    For colorectal tumours, the overall linkage rate between CCR and the DAD/NACRS was consistently above 96% in all provinces and years. Acute care linkage rates ranged from 91% (Nova Scotia, 2008) to 99% (Prince Edward Island, 2005). Day-surgery linkage rates were similar across years and provinces, ranging from 68% (Nova Scotia, 2005) to 79% (Prince Edward Island, 2007). Men had consistently higher linkage rates than did women (data not shown). For example, in Manitoba, ranges were from 72% to 79% for men and from 64% to 73% for women.

    For prostate cancer, overall linkage rates to hospitalizations varied across provinces from 75% to 80% in Manitoba to more than 90% in Ontario. Linkage rates to acute care hospitalizations ranged from 61% (Ontario, Nova Scotia, 2007) to 72% (Prince Edward Island, 2005). Linkage rates to day surgery ranged from less than 40% in Manitoba to more than 60% in Nova Scotia.

    As noted, linkage rates to oncology-related services in the NACRS were available only for Ontario (Table 3). For breast tumours, rates ranged from 81% to 92%. For colorectal tumours, rates were around 64%, for rectum/rectosigmoid junction tumours, around 75% (data not shown), for all years except 2008. For prostate tumours, linkage rates to oncology-related services ranged from 42% to 56%.

    Treatment rates

    Breast tumours
    Overall, more than 90% of cancer tumours identified between 2005 and 2008 were linked to at least one type of treatment during the follow-up period, except in Manitoba where the  overall rate was 88.8% in 2008 (Table 4). In Ontario, the overall treatment rate ranged between 92.1% and 93.3% when all DAD and NACRS hospital events were considered, compared with 90.8% to 91.9% when only acute and day surgery admissions were considered. Overall treatment rates varied with the age of the patient and the stage of the breast tumour at diagnosis. Tumours among patients aged 70 or older were generally less likely than tumours among younger women to link to at least one treatment (Appendix B). Information about tumour staging was available only for Manitoba and Prince Edward Island. In both provinces, linkage rates to any of the selected treatments were lowest for stage IV tumours (Appendix C).

    Treatment rates were generally highest for breast-conserving surgery and/or mastectomy in all provinces. The overall rate of treatment ranged from 89.1% to 90.5% in Ontario, from 86.6% to 91% in Manitoba, from 75.6% to 90.1% in Nova Scotia, and from 87.7% to 92.7% in Prince Edward Island. When specific procedures were considered, over half of tumours received breast-conserving surgery in Ontario and Manitoba, compared with about 35% in Nova Scotia and Prince Edward Island. Conversely, around one in three breast tumours received mastectomy only in Ontario and Manitoba, compared with approximately 40% in Nova Scotia, and between 33% and 43% in Prince Edward Island.

    The rate of lymph node treatment varied across provinces and years. In Ontario, Nova Scotia and Prince Edward Island, the percentage of breast cancer tumours with at least one lymph node treatment was 80% or higher, except for Nova Scotia in 2008 (67.4%). The rate for Manitoba varied from 73.1% (2005) to 76.9% (2006). These rates may not have captured removal of nodes that occurred in radical excisions of breast (mastectomy); thus, numbers do not represent absolute rates of lymph node excisions.

    Treatment rates for radiation were highest in Ontario―about 60%―when all hospital events were considered. The Ontario rate fell to less than 1% (similar to other provinces) when radiation treatment rates were restricted to acute care and day surgery hospital events.

    Treatment rates for diagnostic biopsy (breast, not otherwise specified) were highest in Nova Scotia, ranging from 43.6% to 55.7%. This compared with approximately 5% in Ontario and less than 5% in Manitoba.

    Differences in treatment rates also emerged by type of admission: acute versus day surgery. In Ontario, for example, linkage rates to same-day surgery exceeded linkage rates to acute care treatment (data not shown), owing largely to a greater likelihood of linking to “breast-conserving surgery only” in day surgery. In Manitoba and Prince Edward Island, linkage rates to treatments in acute care settings were generally higher than to day surgery (data not shown). In Nova Scotia, linkage rates to at least one treatment in acute care relative to rates for day surgery varied by year.

    Colorectal tumours
    The overall treatment rates for colorectal tumours to any of the treatments was 90% or higher across all provinces and years (Table 5). Treatment rates for colorectal tumours varied by sex, age and tumour stage. Rates tended to be lower for women than for men (Appendix D) and for those aged 70 or older (Appendix E). For the two provinces where tumour stage information was available, consistent patterns in treatment rates emerged, with rates lowest among those with Stage IV (Appendix F).

    Colorectal treatment rates primarily reflect surgery. Surgical treatment included excisions in the colon, and/or the rectum/rectosigmoid junction, pelvic or intra-abdominal lymph nodes, and tended to occur in acute care settings (data not shown). This explains why, for this cancer, the broader array of data for Ontario did not yield much higher overall colorectal treatment rates than those in the other provinces.

    Treatment rates were generally highest for surgery/excision, with 80% or more of colorectal tumours experiencing this treatment. The surgical treatment rates were consistently highest in Ontario (overall rate) and Prince Edward Island.

    More than 70% of colorectal tumours in Ontario, Manitoba and Prince Edward Island had a linked diagnostic procedure with at least one biopsy and/or colonoscopy, proctoscopy, lymph node excision. The corresponding diagnostic treatment rate in Nova Scotia was consistently below 70%.

    Rates of treatments to repair or reconstruct the large intestine and/or rectum varied from about 5% in Ontario and Nova Scotia to 7% to 9% in Manitoba and Prince Edward Island.

    Between 10% and 12 % of colorectal tumours in Ontario linked to radiation treatments when all available services were considered. However, based on combined acute care and day surgery, the treatment rate fell to 1% or less rates, which was comparable to radiation treatment rates in the other three provinces. The pattern was similar for pharmacotherapy treatment rates. At least one-third of colorectal tumours in Ontario linked to a pharmacotherapy treatment when all hospital events were considered, but the figure decreased to 2% when acute and day surgery hospital events were considered. Linkage rates in Manitoba, Nova Scotia, and Prince Edward Island were 2% or less.

    Differences in treatment rates were also noted by type of service—acute versus day surgery. Overall treatment rates were higher in acute care, than in day surgery in all provinces, primarily due to higher surgical rates in acute care (data not shown). The reverse was true for the diagnostic treatment category, where rates were higher in day surgery than in acute care in all provinces.

    Colon and rectum/rectosigmoid junction tumours
    Treatment rates for single malignant colon tumours resembled patterns for colorectal treatment (Table 6). This was expected, given that colon tumours comprised the majority (63% to 81%) of single malignant colorectal tumours.

    However, treatment rates for surgery for colon tumours tended to be slightly higher than rates for colorectal tumours. At least 80% of these tumours linked to surgical treatments in each province. As well, 60% to 77% of these tumours linked to diagnostic treatments.

    As expected, in Ontario, radiation treatment rates of colon tumours were high when all hospital events were considered, but similar to those of other provinces based only on combined acute care and day surgery:1% to 2%, compared with less than 1% in the other provinces. Rates for systemic pharmacotherapy treatment in Ontario were at least 27% when all service types were considered, but only about 2% based on combined day surgery and acute care hospitalizations.

    Based on all hospital service types, Ontario treatment rates for radiation and systemic pharmacotherapy were higher for rectum/rectosigmoid junction than for colon tumours (Table 7). Treatment rates for repairs also tended to be higher for rectum/rectosigmoid junction tumours, compared with colon tumours (Tables 6 and 7). Repair treatments were more likely to occur as day surgery than within acute care (data not shown).

    Prostate tumours
    The overall treatment rate for prostate cancer varied across provinces. In Ontario, about three-quarters of tumours linked to at least one treatment when all service types were considered; the rate fell to about 50% when only inpatient and day surgery admissions were considered (Table 8). Comparable rates ranged from 42% to 45% in Manitoba, from 58% to 71% in Nova Scotia, and from 43% to 53% in Prince Edward Island. Consistent with other cancers, treatment rates for prostate tumours were inversely associated with patient age at diagnosis, with the lowest rates among men aged 70 or older (Appendix G). Patterns by stage were u-shaped; stage III tumours tended to have highest overall treatment rates (Appendix H).

    Surgery was the most prevalent treatment in all provinces, with about 40% of prostate cancer tumours linking to at least one type of surgical intervention. In Ontario and Manitoba, 35% to 40% of tumours linked to a surgical treatment; in Prince Edward Island, 25% to 37%; and in Nova Scotia, from 40% to 43%. In every year, surgical treatment rates were highest in Nova Scotia. The interventions primarily contributing to these rates were radical excisions to, or complete destruction of, the prostate. These treatments generally were more likely to occur during acute care hospitalizations than in day surgery (data not shown) in all provinces except Nova Scotia.

    Treatment rates for diagnostic procedures for prostate tumours varied widely across provinces. Rates were highest in Nova Scotia (30% to 47%), followed by Ontario (24% to 28%, overall rate), Manitoba (5% to 9%), and Prince Edward Island (4% or less). The diagnostic procedure that primarily accounted for these rates was biopsy of the prostate. Diagnostic and surgical rates mainly accounted for differences across provinces for overall treatment rates.

    As with other cancers, radiation treatment rates were highest in Ontario where about one-third of prostate tumours linked to radiation (29% to 33%); these rates fell to about 3% when acute care and day surgery only were considered. Comparable radiation rates were highest (9% to 18%) in Prince Edward Island; in most cases, these represented brachytherapy interventions submitted by hospitals in New Brunswick. Radiation rates ranged from 2% to 6% in Nova Scotia, and were not reportable in Manitoba because of small numbers.

    Treatment rates for systemic pharmacotherapy for prostate cancer were about 5% in Ontario based on all service types, and about 1% based on combined day surgery and acute care services. Almost no prostate tumours in the other provinces linked to systemic pharmacotherapy.

    Date modified: