The bowel cancer Atlas domain presents access, quality and outcome indicators across district health boards and Regional Cancer Network regions for people diagnosed with bowel cancer (adenocarcinoma of the colon or rectum).
The goal of the Atlas is to explore whether there are wide variations between DHBs in the incidence, treatment and, where possible, outcomes in particular disease areas. The Atlas is designed to prompt debate and raise questions about why differences exist and to stimulate improvement through this debate. This can act as a trigger to spur local action to determine the causes and, if appropriate, local quality improvement activities to improve the quality of care and equity of health care services and outcomes for New Zealanders.
|Bowel cancer single map||Bowel cancer consumer summary|
- Twenty three percent of people had localised disease spread, and 21 percent had distant disease spread at the time of their bowel cancer diagnosis. About a quarter of people with bowel cancer were diagnosed following a visit to an emergency department (ED).
- Two-thirds of people with bowel cancer had surgery in a public hospital
- On average people stayed in a public hospital for 10 days after surgery
- Five percent of people died within three months of surgery in public hospitals
- More than two-thirds of people were alive two years after being diagnosed with bowel cancer
- There was wide variation between DHBs in the use of radiotherapy before surgery for people with rectal cancer
- There was up to two-fold variation between DHBs for:
- mortality at three months after surgery
- surgery which usually results in a permanent colostomy for people with rectal cancer
- Many of the indicators showed variation between ethnic groups and/or age groups. There was wide variation between DHBs for presentation rates at an emergency department (ED) prior to bowel cancer diagnosis for Māori.
Key findings from the 2009–13 data
People diagnosed with bowel cancer in New Zealand
Between 2009 and 2013, 14,394 people living in DHB areas were registered as having bowel cancer. Of these, 9894 people were registered as having colon cancer, 1008 as having cancer of the rectosigmoid junction and 3767 as having rectal cancer. Some people had cancer in more than one of these sites.
Some indicators are presented for bowel cancer overall and some for people with rectal cancer only. Methods used to calculate these indicators are available here. A summary of the cancer pathway and indicators developed at each step is shown in the figure 1.
Figure 1. Bowel cancer treatment pathway and indicators in the bowel cancer domain of the Atlas of Healthcare Variation
Disease extent is recorded on the Cancer Registry as follows:
- Localised – cancer is limited to the place where it started, with no sign it has spread.
- Adjacent – cancer has directly spread to nearby structures (excluding lymph nodes).
- Regional – cancer has spread to regional lymph nodes.
- Distant – cancer has spread to distant parts of the body.
- Unknown – there is not enough information to figure out the stage.
Commentary is only provided when there are significant differences between DHBs or ethnic, age or gender groups. In some instances there were too few people to examine differences between ethnic groups.
- Patterns for treatment indicators will be influenced by the absence of private hospital treatment data e.g. surgery rates will be undercounted for some DHBs. People who had part of their treatment in a private hospital and part in public hospitals will also be missing from these analyses. Grey shading in the table indicates where all public hospital but limited private hospital data is available.
- In the Atlas people are included in the DHB where they usually live. This is not always the DHB that provided treatment.
Bowel cancer diagnosis
The overall unadjusted population rate for people diagnosed with bowel cancer was 65.7 per 100,000 people, with DHB rates ranging from 44.4 to 113.0 per 100,000. In addition, the rates were adjusted to take into account the different age distributions in DHBs. The age adjusted bowel cancer rates show less variation between DHBs.
Māori and Pacific peoples had a higher proportion of distant disease at diagnosis
Disease extent is one of the most important factors in planning how to treat bowel cancer and determining how successful treatment might be.
- Twenty-three percent of people had localised disease at the time of diagnosis, 21 percent had distant disease and for 18 percent of people disease extent was not available on the Cancer Registry.
- Māori and Pacific peoples had higher proportions (28 and 25 percent, respectively) with distant disease than other ethnic groups (who ranged between 17 and 21 percent).
Presentation rates at ED prior to bowel cancer diagnosis for Māori showed wide DHB variation
Patients diagnosed with bowel cancer following ED presentation are managed differently to those referred via other pathways. Variation in this indicator may reflect variations in access to primary and secondary care.
- On average, 27 percent of people were diagnosed following ED presentation at a public hospital with two fold variation between DHBs.
- People aged 80 years and over (33 percent) and under 40 years (40 percent) had higher ED presentation rates before diagnosis with bowel cancer.
- More Māori (39 percent) and Pacific peoples (41 percent) were diagnosed with bowel cancer following ED presentation. There was 2.5 fold variation between DHBs for ED presentation for Māori (23–58 percent).
- The ED presentation rate for females (29 percent) was higher than for males (26 percent).
Bowel cancer surgery in public hospitals
People in their 70s and under 40 years of age with bowel cancer had the highest public hospital surgical resection rates
This indicator defines the population for many of the other treatment indicators. Patterns for this indicator will be strongly influenced by the absence of private hospital use data.
Removal of a locally confined cancer by surgery is the most certain method of curing bowel cancer but patient suitability and disease characteristics have a major influence on treatment.
- Sixty-three percent of people with bowel cancer had curative surgery in a public hospital.
- Higher resection (surgery) rates were seen for people aged 70–79 years (69 percent) and under 40 years of age (66 percent) than other age groups.
Māori, Pacific peoples and males had longer public hospitals stays after surgery for bowel cancer
The length of time people with bowel cancer stay in hospital after surgery varies according to their state of health before and after the operation, the extent of their cancer, the type of operation they have, any problems associated with that procedure and the social care they have available at home.
Long stays in hospital after surgery may not only be detrimental to patients (eg, greater risk of exposure to infection) but are also expensive for hospitals.
- The overall average length of stay for patients with bowel cancer following surgery was 10 days. This ranged from 7.1 days to 11.9 days across DHBs.
- On average, the length of stay for males (11 days) was one day longer than for females (10 days).
- The average length of stay for Māori and Pacific peoples (11 days) was longer than for other ethnic groups (10 days).
There was two-fold variation among DHBs in 90-day mortality rates for people with bowel cancer after curative surgery in public hospitals
While surgical removal of a tumour has many benefits, people with bowel cancer may experience serious potential complications including risk to life because of surgery. The risk of dying after surgery can be influenced by the extent of the cancer, the other medical conditions a patient suffers from, and the number of cancers treated by the surgical unit. Improvements in surgical techniques and care of patients are likely, over time, to lead to better survival rates for people with bowel cancer who have surgery.
- Five percent of people died within 90 days of undergoing curative surgery. There was two-fold variation across DHBs in 90-day mortality rates (3–8 percent) with one DHB at 13 percent.
- The 90-day mortality rate for people 80 years and over (10 percent) was twice that of other age groups (2–5 percent).
- The over 80-year-old mortality rate varied across DHBs from 7 percent to 17 percent.
Chemotherapy in public hospitals
Chemotherapy rates in public hospitals decreased in relation to increasing age in people with localised and adjacent bowel cancer
In localised colon cancer, surgical resection alone is widely accepted as standard treatment, and additional therapy is rarely needed. Most people with adjacent disease who undergo surgery will be cancer-free five years later, without chemotherapy.
- Four percent of people with localised extent and 16 percent with adjacent extent bowel cancer were treated with chemotherapy.
- Chemotherapy rates decreased with age for:
- localised bowel cancer (from 8 percent in under 40-year-olds to 1 percent in people aged 80 years and older)
- adjacent bowel cancer (from 50 percent in under 40-years-olds to 2 percent in people aged 80 years and older).
- DHB chemotherapy rates varied from 4 percent to 11 percent for localised disease extent.
Females and older people had lower chemotherapy rates in public hospitals for regional extent bowel cancer
The standard treatment for people with regional extent disease is surgery followed by chemotherapy (called adjuvant chemotherapy, meaning it is given after surgery to reduce the risk of a recurrence of the cancer). People who aren’t healthy enough for surgery may still have radiation therapy and/or chemotherapy.
- Overall, 62 percent of people with regional extent bowel cancer were treated with chemotherapy.
- Chemotherapy rates for regional extent disease decreased with age, from 92 percent aged under 40 years to 12 percent in people aged 80 years and older.
- Males with regional extent disease (66 percent) had a higher chemotherapy rate than females (58 percent).
Chemotherapy rates in public hospitals decreased in relation to increasing age in people with distant extent bowel cancer
Chemotherapy is typically given to people with distant extent disease before and/or after surgery or instead of surgery. If the metastases (cancer that has spread to distant parts of the body) cannot be removed because they are too large or there are too many of them, chemotherapy may be given before any surgery (neoadjuvant chemotherapy). Then, if the tumours shrink, surgery to remove them may be attempted. Chemotherapy would then be given again after surgery.
- Forty-eight percent of people with distant extent bowel cancer were treated with chemotherapy.
- Chemotherapy rates for distant extent disease decreased with age from 87 percent in people aged under 40 years to 10 percent in people aged 80 years and older.
Rectal cancer treatment in public hospitals
Treatment options for people with rectal cancer are complex and variable compared with the treatment options for people with colon cancer.
Rectal cancer is more prone to local recurrence as it is treated within the confines of the pelvis. The challenge is to achieve a good functional outcome for the patient and reduce the risk of the cancer recurring in the same area.
People in their 70s had the highest surgical resection rates in public hospitals for rectal cancer
- Fifty-seven percent (2156) of people with rectal cancer had surgery in a public hospital within one year of diagnosis.
- Resection rates varied from 45 percent to 71 percent across DHBs.
- People in the 70–79-year age group had the highest resection rates (63 percent) and people aged 80 and over had the lowest rates (48 percent).
Rate of chemotherapy in conjunction with long-course radiotherapy for people with rectal cancer decreased with increasing age
Long-course radiotherapy (20 or more sessions before or after surgery) provides better local control of rectal cancer if given in conjunction with chemotherapy.
- Eighty-five percent (790) of the 886 people who were treated with surgery and long course radiotherapy for rectal cancer had chemotherapy at the same time.
- The proportion of patients with rectal cancer receiving long-course radiotherapy, who also had chemotherapy, decreased with age from 96 percent those aged under 40 years to 59 percent in those aged 80 years and over.
There was two-fold DHB variation between DHBs in radiotherapy before surgery rates in public hospitals for people with rectal cancer
Radiotherapy can reduce the risk of recurrence of some rectal cancers. It is preferable to give radiotherapy before surgery, as it causes fewer side effects compared with radiotherapy after surgery.
- Fifty-five percent of people with rectal cancer had radiotherapy before surgery with DHB variation ranging from 35 percent to 69 percent.
- The percentage of people with rectal cancer receiving radiotherapy before surgery decreased with age, from 64 percent for people aged under 40 years, to 41 percent for people aged 80 years and over.
There was wide variation across DHBs in the use of short-course radiotherapy in public hospitals, for people with rectal cancer
Pre-operative radiotherapy can be given as short-course radiotherapy (SCRT) or as long-course radiotherapy (LCRT). SCRT has fewer short-term side effects than LCRT, is more convenient for the patient and requires fewer resources. However there are some concerns that SCRT may have more long-term side effects and may be less effective for more advanced tumours. SCRT also requires more coordination, as surgery should occur within a week of completing it. Clinical trials comparing the two approaches are ongoing. However guidance derived from current evidence is available to help clinicians and patients choose between these two treatment options.
- Overall, 55 percent (827) of people with rectal cancer undergoing surgery had preoperative radiotherapy and 33 percent (363 people) had SCRT.
- There was wide variation (13–82 percent) between DHBs in the proportion of people receiving pre-operative radiotherapy who received SCRT.
- A higher proportion of women (35 percent) receiving pre-operative radiotherapy had SCRT compared with men (28 percent).
- People aged over 80 years and over were twice as likely as younger people to receive SCRT.
There was two-fold variation across DHBs in rates of surgery that usually result in a permanent colostomy people with rectal cancer
Low rectal surgery often results in a permanent colostomy. A colostomy has a significant effect on the patient following surgery and may be associated with higher morbidity and poorer quality of life for patients. Advances in this area and specialist centres mean the outcomes for people with rectal cancer have potential to improve.
- Thirty-three percent (713) of people with rectal cancer had a Hartman's a Hartmann’s procedure or abdominoperineal resection in a public hospital. These procedures usually result in a permanent colostomy
- Rates for surgery that usually results in a permanent colostomy varied from 23 percent to 56 percent across DHBs.
Survival for people with bowel cancer
For most people with bowel cancer, survival and cure remain their primary concern after diagnosis.
Māori and people aged 80 years and over had higher mortality rates two years after diagnosis
Variation in two-year mortality rates is likely to reflect, at least in part, differences in the quality of surgery, patient characteristics and whether the patient had access to appropriate chemotherapy and radiotherapy.
- Sixty-nine percent of people diagnosed with bowel cancer were alive two years after their diagnosis.
- The two-year mortality rate for people aged 80 years and over (48 percent) was higher than for people of other ages (22–29 percent).
- A higher proportion of Māori (38 percent) died within two years of diagnosis than people of Asian and European ethnicity (17–32 percent).
What questions do these indicators prompt?
- To what extent do treatment rates reflect patient characteristics or availability of health system resources or something else?
- What effect does the lack of private hospital data have on the treatment rates presented here?
- To what extent do low surgical rates in some DHBs reflect missing private hospital data?
- Are people with rectal cancer in different DHBs offered the best radiotherapy/chemotherapy treatment options?
- How does the quality of public hospital services influence treatment options and survival for bowel cancer patients?
- Are the outcomes for people with bowel cancer treated in private hospitals better or worse than those for people treated in public hospitals?
- Why do ethnic groups and/or age group rates differ for many of the indicators?
Early diagnosis of symptomatic cancer is likely to lead to improved survival, earlier stage diagnosis and improved quality of life.
Bowel cancer is the most commonly reported cancer in New Zealand, with approximately 3000 cases and 1200 deaths each year.
Faster cancer treatment (FCT) indicators were introduced by the Ministry of Health in July 2012. The indicators require DHBs to collect standardised information on patients who had been referred urgently with a high suspicion of cancer. These FCT indicators focus only on people who are being treated for bowel cancer.
The recently published PIPER study noted that, while national tumour standards provide a focus on better care for patients, a gap exists in the monitoring and reporting of outcomes. The study also noted the need to identify those who are missing out on treatment.
The provisional Standards of Service Provision for Bowel Cancer Patients in New Zealand provided a helpful starting point for consideration of indicators for an Atlas domain.
This project was carried out with the assistance of an Expert Advisory Group that included members of the National Bowel Cancer Working Group and other bowel cancer experts in New Zealand.
A pilot bowel cancer screening programme has been running in the Waitemata DHB region from late 2011. This Atlas domain will provide a useful baseline for assessing improvements in patient outcomes when this pilot is extended and a nationwide bowel cancer screening programme is implemented throughout New Zealand.
The indicators in this Atlas include all people diagnosed with bowel cancer and use data from the New Zealand Cancer Registry, the Pharmaceutical Collection and national admitted patients and outpatient databases. The Cancer Registry includes patients diagnosed with cancer in both public and private hospitals. However treatment data (surgery, radiotherapy and chemotherapy) is available in New Zealand National Collections for very few private hospitals. Previous studies have found that National Collections undercounted receipt of surgery by 13–19 percent, and receipt of chemotherapy or radiotherapy for breast cancer patients by 18 percent and 16 percent respectively.
While the findings in the Atlas may be affected by missing private hospital data the results are generally very similar to those recently reported by the Piper study. The Piper Report was a large study of bowel cancer patients and their treatment, and included data from public and private hospitals in New Zealand.
A note is made in the Atlas commentary where the treatment rates may be underestimated by missing private hospital data.
Further information about health services for people with bowel cancer in New Zealand can be found on the Ministry of Health website: http://www.health.govt.nz/your-health/conditions-and-treatments/diseases-and-illnesses/bowel-cancer.
- Gurney J, et al. 2013. The completeness of cancer treatment data on the National Health Collections. NZ Med J 126(1381): 69–74.